Instructions to use Zipeng365/WISP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Scikit-learn
How to use Zipeng365/WISP with Scikit-learn:
from huggingface_hub import hf_hub_download import joblib model = joblib.load( hf_hub_download("Zipeng365/WISP", "sklearn_model.joblib") ) # only load pickle files from sources you trust # read more about it here https://skops.readthedocs.io/en/stable/persistence.html - Notebooks
- Google Colab
- Kaggle
Refresh bilingual research videos with MiMo narration and synchronized subtitles
Browse files- .gitattributes +8 -0
- README.md +12 -8
- videos/REPRODUCE.md +26 -0
- videos/WISP_MiMo_reproducible_project.zip +3 -0
- videos/WISP_en.srt +129 -117
- videos/WISP_en.vtt +117 -108
- videos/WISP_en_mimo_voice.mp4 +3 -0
- videos/WISP_en_mimo_voice.wav +3 -0
- videos/WISP_en_script.txt +36 -59
- videos/WISP_en_system_voice.mp4 +2 -2
- videos/WISP_zh.srt +76 -72
- videos/WISP_zh.vtt +52 -49
- videos/WISP_zh_mimo_voice.mp4 +3 -0
- videos/WISP_zh_mimo_voice.wav +3 -0
- videos/WISP_zh_script.txt +22 -32
- videos/WISP_zh_system_voice.mp4 +2 -2
- videos/archive/20261023/WISP_en.srt +235 -0
- videos/archive/20261023/WISP_en.vtt +178 -0
- videos/archive/20261023/WISP_en_cover.png +3 -0
- videos/archive/20261023/WISP_en_script.txt +82 -0
- videos/archive/20261023/WISP_en_system_voice.mp4 +3 -0
- videos/archive/20261023/WISP_zh.srt +183 -0
- videos/archive/20261023/WISP_zh.vtt +139 -0
- videos/archive/20261023/WISP_zh_cover.png +3 -0
- videos/archive/20261023/WISP_zh_script.txt +69 -0
- videos/archive/20261023/WISP_zh_system_voice.mp4 +3 -0
- videos/release_manifest.json +62 -0
- videos/voice_release_20261028.json +17 -0
.gitattributes
CHANGED
|
@@ -37,3 +37,11 @@ videos/WISP_en_system_voice.mp4 filter=lfs diff=lfs merge=lfs -text
|
|
| 37 |
videos/WISP_en_cover.png filter=lfs diff=lfs merge=lfs -text
|
| 38 |
videos/WISP_zh_system_voice.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 39 |
videos/WISP_zh_cover.png filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
videos/WISP_en_cover.png filter=lfs diff=lfs merge=lfs -text
|
| 38 |
videos/WISP_zh_system_voice.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 39 |
videos/WISP_zh_cover.png filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
videos/archive/20261023/WISP_en_cover.png filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
videos/archive/20261023/WISP_en_system_voice.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
videos/archive/20261023/WISP_zh_cover.png filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
videos/archive/20261023/WISP_zh_system_voice.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
videos/WISP_zh_mimo_voice.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
videos/WISP_zh_mimo_voice.wav filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
videos/WISP_en_mimo_voice.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
videos/WISP_en_mimo_voice.wav filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -30,22 +30,26 @@ Benchmark data remain in [WristHARBench on Hugging Face](https://huggingface.co/
|
|
| 30 |
|
| 31 |
Two independently narrated animations explain WristHARBench, how WISP representations read wrist-motion signals, how complementary information is combined, and the experimental findings.
|
| 32 |
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
|
| 36 |
-
|
|
|
|
|
|
|
|
| 37 |
|
| 38 |
### English
|
| 39 |
|
| 40 |
-
<video controls preload="none" width="100%" poster="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_en_cover.png" src="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/
|
| 41 |
|
| 42 |
### 中文
|
| 43 |
|
| 44 |
-
<video controls preload="none" width="100%" poster="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_zh_cover.png" src="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/
|
|
|
|
|
|
|
| 45 |
|
| 46 |
-
|
| 47 |
|
| 48 |
-
[
|
| 49 |
|
| 50 |
## Install
|
| 51 |
|
|
|
|
| 30 |
|
| 31 |
Two independently narrated animations explain WristHARBench, how WISP representations read wrist-motion signals, how complementary information is combined, and the experimental findings.
|
| 32 |
|
| 33 |
+
**Narration updated on 28 October 2026:** Chinese uses MiMo's 茉莉 voice; English uses Chloe. Both are AI-generated voices. Subtitles and scene timing have been rebuilt for the new narration.
|
| 34 |
+
|
| 35 |
+
| Language | Video | Audio | Subtitles | Script |
|
| 36 |
+
| --- | --- | --- | --- | --- |
|
| 37 |
+
| English · 3:46 | [Watch / 下载 MP4](https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_en_mimo_voice.mp4) | [WAV](https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_en_mimo_voice.wav) | [SRT](videos/WISP_en.srt) · [VTT](videos/WISP_en.vtt) | [English script](videos/WISP_en_script.txt) |
|
| 38 |
+
| 中文 · 3:47 | [Watch / 下载 MP4](https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_zh_mimo_voice.mp4) | [WAV](https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_zh_mimo_voice.wav) | [SRT](videos/WISP_zh.srt) · [VTT](videos/WISP_zh.vtt) | [中文脚本](videos/WISP_zh_script.txt) |
|
| 39 |
|
| 40 |
### English
|
| 41 |
|
| 42 |
+
<video controls preload="none" width="100%" poster="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_en_cover.png" src="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_en_mimo_voice.mp4"></video>
|
| 43 |
|
| 44 |
### 中文
|
| 45 |
|
| 46 |
+
<video controls preload="none" width="100%" poster="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_zh_cover.png" src="https://huggingface.co/Zipeng365/WISP/resolve/main/videos/WISP_zh_mimo_voice.mp4"></video>
|
| 47 |
+
|
| 48 |
+
The research story and graphics are retained from the 18 October 2026 production. Signal landscapes include illustrative synthetic inputs, distinguished from experimental results in the videos. This update changes narration and subtitle timing; it does not introduce new experimental results.
|
| 49 |
|
| 50 |
+
[Reproducible project](videos/WISP_MiMo_reproducible_project.zip) · [Reproduction instructions and sources](videos/REPRODUCE.md) · [Paper (Pure)](https://research.birmingham.ac.uk/en/publications/wisp-composable-representations-for-wrist-worn-activity-recogniti/)
|
| 51 |
|
| 52 |
+
[Original system-voice edition and subtitles](https://huggingface.co/Zipeng365/WISP/tree/main/videos/archive/20261023). Previously shared `system_voice.mp4` URLs now serve the refreshed editions.
|
| 53 |
|
| 54 |
## Install
|
| 55 |
|
videos/REPRODUCE.md
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# WISP bilingual narration refresh — 28 October 2026
|
| 2 |
+
|
| 3 |
+
This release preserves the research story and graphics from the 18 October 2026 animation and replaces macOS system speech with MiMo-V2.5-TTS narration (Chinese: 茉莉; English: Chloe). The voices are AI-generated. Pronunciation guidance, the exact narration text, and generated source audio are included.
|
| 4 |
+
|
| 5 |
+
The original release is preserved separately. Public files with the historic `system_voice.mp4` names are compatibility links containing the refreshed MiMo editions.
|
| 6 |
+
|
| 7 |
+
## Reproduce
|
| 8 |
+
|
| 9 |
+
1. Install Python 3.12 and Node.js 24. Run `python -m venv .venv`, then install `requirements.txt` into it.
|
| 10 |
+
2. In `studio`, run `npm ci`. The project uses Remotion 4.0.536. Run `npm ci` in `transcriber` for local Whisper alignment.
|
| 11 |
+
3. The included source audio and transcripts let you rebuild without a speech API. To synthesize again, set `MIMO_API_KEY` locally and run `python synthesize.py --all`. Existing audio is preserved, so move any segment you intend to regenerate to an archive first. MiMo generation is nondeterministic; reruns may differ.
|
| 12 |
+
4. For new audio, run `node transcriber/transcribe.mjs` (configure its Python interpreter path if needed); remove only corresponding old transcript JSON files before rerunning. Existing transcript files are reused.
|
| 13 |
+
5. Run `python engineering/assemble.py`, then `python engineering/register.py`.
|
| 14 |
+
6. Run `python engineering/render.py films`. Set `WISP_CHROME` to your Chrome executable if needed. For interactive playback, run `npx remotion studio` inside `studio`.
|
| 15 |
+
7. Run `python engineering/optimize_media.py`, then `python engineering/check_delivery.py` and review the alignment and media reports in `qa`.
|
| 16 |
+
|
| 17 |
+
Subtitles retain the source script, with timings aligned to local Whisper recognition. ASR spelling errors do not alter the script. The original English CIS subtitle splitting error is corrected. Minor pacing and loudness adjustments are recorded in `qa/alignment.json`.
|
| 18 |
+
|
| 19 |
+
The graphics and research claims were checked on 18 October 2026. This is a narration update, not a new experimental release.
|
| 20 |
+
|
| 21 |
+
## Sources
|
| 22 |
+
|
| 23 |
+
- Speech API: https://mimo.mi.com/docs/en-US/quick-start/usage-guide/audio/speech-synthesis-v2.5
|
| 24 |
+
- Paper: https://research.birmingham.ac.uk/en/publications/wisp-composable-representations-for-wrist-worn-activity-recogniti/
|
| 25 |
+
- Method: https://huggingface.co/Zipeng365/WISP
|
| 26 |
+
- Benchmark: https://huggingface.co/datasets/Zipeng365/WristHARBench
|
videos/WISP_MiMo_reproducible_project.zip
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:797c20c808bf1db672ca298d55c4029630d9ccb16f8f8cf78e99869052349a96
|
| 3 |
+
size 57091240
|
videos/WISP_en.srt
CHANGED
|
@@ -1,235 +1,247 @@
|
|
| 1 |
1
|
| 2 |
-
00:00:
|
| 3 |
-
Walking, cooking, and handling objects
|
| 4 |
|
| 5 |
2
|
| 6 |
-
00:00:
|
| 7 |
-
|
| 8 |
|
| 9 |
3
|
| 10 |
-
00:00:
|
| 11 |
-
Local shape, rhythm, temporal order, and
|
| 12 |
|
| 13 |
4
|
| 14 |
-
00:00:
|
| 15 |
-
|
| 16 |
|
| 17 |
5
|
| 18 |
-
00:00:
|
| 19 |
-
When benchmarks mix sensor locations, they
|
| 20 |
|
| 21 |
6
|
| 22 |
-
00:00:
|
| 23 |
-
|
| 24 |
|
| 25 |
7
|
| 26 |
-
00:00:
|
| 27 |
-
WristHARBench brings together nine wrist
|
| 28 |
|
| 29 |
8
|
| 30 |
-
00:00:
|
| 31 |
-
|
| 32 |
|
| 33 |
9
|
| 34 |
-
00:00:
|
| 35 |
-
Two additional multichannel inertial
|
| 36 |
|
| 37 |
10
|
| 38 |
-
00:00:
|
| 39 |
-
|
| 40 |
|
| 41 |
11
|
| 42 |
-
00:00:
|
| 43 |
-
Within each repeat, different people
|
| 44 |
|
| 45 |
12
|
| 46 |
-
00:00:
|
| 47 |
-
|
| 48 |
|
| 49 |
13
|
| 50 |
-
00:00:
|
| 51 |
-
No person appears in two groups.
|
| 52 |
|
| 53 |
14
|
| 54 |
-
00:00:
|
| 55 |
-
|
| 56 |
|
| 57 |
15
|
| 58 |
-
00:00:
|
| 59 |
-
Methods share the same windows and splits,
|
| 60 |
|
| 61 |
16
|
| 62 |
-
00:00:
|
| 63 |
-
|
| 64 |
|
| 65 |
17
|
| 66 |
-
00:00:
|
| 67 |
-
|
| 68 |
|
| 69 |
18
|
| 70 |
-
00:00:
|
| 71 |
-
|
| 72 |
|
| 73 |
19
|
| 74 |
-
00:
|
| 75 |
-
|
| 76 |
|
| 77 |
20
|
| 78 |
-
00:01:
|
| 79 |
-
|
| 80 |
|
| 81 |
21
|
| 82 |
-
00:01:
|
| 83 |
-
|
| 84 |
|
| 85 |
22
|
| 86 |
-
00:01:
|
| 87 |
-
|
| 88 |
|
| 89 |
23
|
| 90 |
-
00:01:
|
| 91 |
-
|
| 92 |
|
| 93 |
24
|
| 94 |
-
00:01:
|
| 95 |
-
|
| 96 |
|
| 97 |
25
|
| 98 |
-
00:01:
|
| 99 |
-
|
| 100 |
|
| 101 |
26
|
| 102 |
-
00:01:
|
| 103 |
-
values to relative,
|
| 104 |
|
| 105 |
27
|
| 106 |
-
00:01:
|
| 107 |
-
|
| 108 |
|
| 109 |
28
|
| 110 |
-
00:01:
|
| 111 |
-
|
| 112 |
|
| 113 |
29
|
| 114 |
-
00:01:
|
| 115 |
-
|
| 116 |
|
| 117 |
30
|
| 118 |
-
00:01:
|
| 119 |
-
|
| 120 |
|
| 121 |
31
|
| 122 |
-
00:01:
|
| 123 |
-
|
| 124 |
|
| 125 |
32
|
| 126 |
-
00:01:
|
| 127 |
-
|
| 128 |
|
| 129 |
33
|
| 130 |
-
00:01:
|
| 131 |
-
|
| 132 |
|
| 133 |
34
|
| 134 |
-
00:
|
| 135 |
-
|
| 136 |
|
| 137 |
35
|
| 138 |
-
00:
|
| 139 |
-
|
| 140 |
|
| 141 |
36
|
| 142 |
-
00:
|
| 143 |
-
features
|
| 144 |
|
| 145 |
37
|
| 146 |
-
00:
|
| 147 |
-
|
| 148 |
|
| 149 |
38
|
| 150 |
-
00:02:
|
| 151 |
-
|
| 152 |
|
| 153 |
39
|
| 154 |
-
00:02:
|
| 155 |
-
|
| 156 |
|
| 157 |
40
|
| 158 |
-
00:02:
|
| 159 |
-
|
| 160 |
|
| 161 |
41
|
| 162 |
-
00:02:
|
| 163 |
-
|
| 164 |
|
| 165 |
42
|
| 166 |
-
00:02:
|
| 167 |
-
|
| 168 |
|
| 169 |
43
|
| 170 |
-
00:02:
|
| 171 |
-
|
| 172 |
|
| 173 |
44
|
| 174 |
-
00:02:
|
| 175 |
-
|
| 176 |
|
| 177 |
45
|
| 178 |
-
00:02:
|
| 179 |
-
|
| 180 |
|
| 181 |
46
|
| 182 |
-
00:02:
|
| 183 |
-
|
| 184 |
|
| 185 |
47
|
| 186 |
-
00:02:
|
| 187 |
-
|
| 188 |
|
| 189 |
48
|
| 190 |
-
00:02:
|
| 191 |
-
|
| 192 |
|
| 193 |
49
|
| 194 |
-
00:
|
| 195 |
-
|
| 196 |
|
| 197 |
50
|
| 198 |
-
00:
|
| 199 |
-
|
| 200 |
|
| 201 |
51
|
| 202 |
-
00:
|
| 203 |
-
|
| 204 |
|
| 205 |
52
|
| 206 |
-
00:
|
| 207 |
-
|
| 208 |
|
| 209 |
53
|
| 210 |
-
00:
|
| 211 |
-
|
| 212 |
|
| 213 |
54
|
| 214 |
-
00:
|
| 215 |
-
|
| 216 |
|
| 217 |
55
|
| 218 |
-
00:03:
|
| 219 |
-
|
| 220 |
|
| 221 |
56
|
| 222 |
-
00:03:
|
| 223 |
-
|
| 224 |
|
| 225 |
57
|
| 226 |
-
00:03:
|
| 227 |
-
|
| 228 |
|
| 229 |
58
|
| 230 |
-
00:03:
|
| 231 |
-
|
| 232 |
|
| 233 |
59
|
| 234 |
-
00:03:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 235 |
Explore the public releases.
|
|
|
|
| 1 |
1
|
| 2 |
+
00:00:01,004 --> 00:00:04,350
|
| 3 |
+
Walking, cooking, and handling objects leave
|
| 4 |
|
| 5 |
2
|
| 6 |
+
00:00:04,350 --> 00:00:06,715
|
| 7 |
+
different clues in wrist motion.
|
| 8 |
|
| 9 |
3
|
| 10 |
+
00:00:06,715 --> 00:00:12,042
|
| 11 |
+
Local shape, rhythm, temporal order, and relationships between axes can
|
| 12 |
|
| 13 |
4
|
| 14 |
+
00:00:12,042 --> 00:00:13,812
|
| 15 |
+
all help identify an activity.
|
| 16 |
|
| 17 |
5
|
| 18 |
+
00:00:15,168 --> 00:00:18,956
|
| 19 |
+
When benchmarks mix sensor locations, they cannot isolate
|
| 20 |
|
| 21 |
6
|
| 22 |
+
00:00:18,956 --> 00:00:21,022
|
| 23 |
+
what the wrist alone reveals.
|
| 24 |
|
| 25 |
7
|
| 26 |
+
00:00:21,022 --> 00:00:25,264
|
| 27 |
+
WristHARBench brings together nine wrist acceleration datasets
|
| 28 |
|
| 29 |
8
|
| 30 |
+
00:00:25,264 --> 00:00:27,841
|
| 31 |
+
under one evaluation contract.
|
| 32 |
|
| 33 |
9
|
| 34 |
+
00:00:27,841 --> 00:00:31,187
|
| 35 |
+
Two additional multichannel inertial datasets support a
|
| 36 |
|
| 37 |
10
|
| 38 |
+
00:00:31,187 --> 00:00:32,398
|
| 39 |
+
separate sensitivity analysis.
|
| 40 |
|
| 41 |
11
|
| 42 |
+
00:00:34,064 --> 00:00:37,718
|
| 43 |
+
Within each repeat, different people enter training,
|
| 44 |
|
| 45 |
12
|
| 46 |
+
00:00:37,718 --> 00:00:39,929
|
| 47 |
+
validation, and test groups.
|
| 48 |
|
| 49 |
13
|
| 50 |
+
00:00:39,929 --> 00:00:43,814
|
| 51 |
+
No person appears in two groups. Validation guides model
|
| 52 |
|
| 53 |
14
|
| 54 |
+
00:00:43,814 --> 00:00:46,679
|
| 55 |
+
choice; testing comes last.
|
| 56 |
|
| 57 |
15
|
| 58 |
+
00:00:46,679 --> 00:00:49,198
|
| 59 |
+
Methods share the same windows and splits,
|
| 60 |
|
| 61 |
16
|
| 62 |
+
00:00:49,198 --> 00:00:51,045
|
| 63 |
+
across five repeated holdouts.
|
| 64 |
|
| 65 |
17
|
| 66 |
+
00:00:52,296 --> 00:00:57,815
|
| 67 |
+
WISP reads the same signal window through complementary representations.
|
| 68 |
|
| 69 |
18
|
| 70 |
+
00:00:57,815 --> 00:01:02,412
|
| 71 |
+
Statistics and spectra describe overall variation and rhythm; other
|
| 72 |
|
| 73 |
19
|
| 74 |
+
00:01:02,412 --> 00:01:06,527
|
| 75 |
+
branches retain shape, order, and cross-axis structure.
|
| 76 |
|
| 77 |
20
|
| 78 |
+
00:01:06,527 --> 00:01:09,642
|
| 79 |
+
These landscapes illustrate operators on synthetic signals.
|
| 80 |
|
| 81 |
21
|
| 82 |
+
00:01:11,141 --> 00:01:14,487
|
| 83 |
+
Random convolution kernels move through time, responding
|
| 84 |
|
| 85 |
22
|
| 86 |
+
00:01:14,487 --> 00:01:16,679
|
| 87 |
+
to local waveform patterns.
|
| 88 |
|
| 89 |
23
|
| 90 |
+
00:01:16,679 --> 00:01:21,391
|
| 91 |
+
A selected slice reveals the responses within one signal window.
|
| 92 |
|
| 93 |
24
|
| 94 |
+
00:01:21,391 --> 00:01:26,737
|
| 95 |
+
Positive proportion, maximum, mean, and standard deviation pool that
|
| 96 |
|
| 97 |
25
|
| 98 |
+
00:01:26,737 --> 00:01:28,301
|
| 99 |
+
response into four features.
|
| 100 |
|
| 101 |
26
|
| 102 |
+
00:01:29,505 --> 00:01:33,775
|
| 103 |
+
The state-and-order representation maps values to relative,
|
| 104 |
|
| 105 |
27
|
| 106 |
+
00:01:33,775 --> 00:01:35,775
|
| 107 |
+
within-window quantile states.
|
| 108 |
|
| 109 |
28
|
| 110 |
+
00:01:35,775 --> 00:01:40,025
|
| 111 |
+
It counts how often states occur and how they transition to each other.
|
| 112 |
|
| 113 |
29
|
| 114 |
+
00:01:40,025 --> 00:01:44,026
|
| 115 |
+
The matrix contains transition probabilities, not attention weights.
|
| 116 |
|
| 117 |
30
|
| 118 |
+
00:01:45,527 --> 00:01:51,546
|
| 119 |
+
Geometry, interval, and shape features combine cross-axis structure,
|
| 120 |
|
| 121 |
31
|
| 122 |
+
00:01:51,546 --> 00:01:54,969
|
| 123 |
+
local statistics, and subsequence matches.
|
| 124 |
|
| 125 |
32
|
| 126 |
+
00:01:54,969 --> 00:01:58,854
|
| 127 |
+
A template moves along the window. The lowest distance
|
| 128 |
|
| 129 |
33
|
| 130 |
+
00:01:58,854 --> 00:02:00,835
|
| 131 |
+
identifies its best match.
|
| 132 |
|
| 133 |
34
|
| 134 |
+
00:02:00,835 --> 00:02:04,058
|
| 135 |
+
Both match quality and location become features.
|
| 136 |
|
| 137 |
35
|
| 138 |
+
00:02:05,271 --> 00:02:09,214
|
| 139 |
+
Complementary clues can be combined at the feature level.
|
| 140 |
|
| 141 |
36
|
| 142 |
+
00:02:09,214 --> 00:02:15,060
|
| 143 |
+
CIS concatenates convolution features R, interval features I, and
|
| 144 |
|
| 145 |
37
|
| 146 |
+
00:02:15,060 --> 00:02:18,771
|
| 147 |
+
spectral, temporal, and cross-axis features S.
|
| 148 |
|
| 149 |
38
|
| 150 |
+
00:02:18,771 --> 00:02:21,368
|
| 151 |
+
A single classifier reads the combined vector
|
| 152 |
|
| 153 |
39
|
| 154 |
+
00:02:21,368 --> 00:02:23,156
|
| 155 |
+
and produces activity scores.
|
| 156 |
|
| 157 |
40
|
| 158 |
+
00:02:24,333 --> 00:02:27,679
|
| 159 |
+
Score ensembles combine independently fitted members after
|
| 160 |
|
| 161 |
41
|
| 162 |
+
00:02:27,679 --> 00:02:29,775
|
| 163 |
+
aligning their activity classes.
|
| 164 |
|
| 165 |
42
|
| 166 |
+
00:02:29,775 --> 00:02:35,910
|
| 167 |
+
CSE averages three members equally; ESE uses weighted averaging.
|
| 168 |
|
| 169 |
43
|
| 170 |
+
00:02:35,910 --> 00:02:39,371
|
| 171 |
+
This is different from concatenating features before one classifier.
|
| 172 |
|
| 173 |
44
|
| 174 |
+
00:02:40,719 --> 00:02:45,508
|
| 175 |
+
On the frozen paper evaluation, WISP with random search reaches
|
| 176 |
|
| 177 |
45
|
| 178 |
+
00:02:45,508 --> 00:02:49,642
|
| 179 |
+
70.20 percent mean macro F1.
|
| 180 |
|
| 181 |
46
|
| 182 |
+
00:02:49,642 --> 00:02:53,412
|
| 183 |
+
RDST, the strongest of eighteen external baselines on this
|
| 184 |
|
| 185 |
47
|
| 186 |
+
00:02:53,412 --> 00:02:57,027
|
| 187 |
+
average, reaches 59.62 percent.
|
| 188 |
|
| 189 |
48
|
| 190 |
+
00:02:57,027 --> 00:03:02,565
|
| 191 |
+
The difference is 10.58 percentage points. This is a dataset average,
|
| 192 |
|
| 193 |
49
|
| 194 |
+
00:03:02,565 --> 00:03:04,331
|
| 195 |
+
not a win on every dataset.
|
| 196 |
|
| 197 |
50
|
| 198 |
+
00:03:05,750 --> 00:03:07,962
|
| 199 |
+
Controlled ablations test complementary
|
| 200 |
|
| 201 |
51
|
| 202 |
+
00:03:07,962 --> 00:03:10,115
|
| 203 |
+
information with a fixed classifier.
|
| 204 |
|
| 205 |
52
|
| 206 |
+
00:03:10,115 --> 00:03:14,481
|
| 207 |
+
With the other two groups already present, adding R gains 4.80
|
| 208 |
|
| 209 |
53
|
| 210 |
+
00:03:14,481 --> 00:03:17,962
|
| 211 |
+
points; adding S gains 3.01.
|
| 212 |
|
| 213 |
54
|
| 214 |
+
00:03:17,962 --> 00:03:22,096
|
| 215 |
+
Adding I changes the mean by minus 0.27 points, with
|
| 216 |
|
| 217 |
55
|
| 218 |
+
00:03:22,096 --> 00:03:24,000
|
| 219 |
+
an interval crossing zero.
|
| 220 |
|
| 221 |
56
|
| 222 |
+
00:03:24,000 --> 00:03:27,308
|
| 223 |
+
The evidence supports adding missing clues, rather than more
|
| 224 |
|
| 225 |
57
|
| 226 |
+
00:03:27,308 --> 00:03:28,173
|
| 227 |
+
overlapping representations.
|
| 228 |
|
| 229 |
58
|
| 230 |
+
00:03:29,658 --> 00:03:34,523
|
| 231 |
+
WristHARBench provides benchmark data. WISP provides method code, saved
|
| 232 |
|
| 233 |
59
|
| 234 |
+
00:03:34,523 --> 00:03:36,965
|
| 235 |
+
results, and retained search models.
|
| 236 |
+
|
| 237 |
+
60
|
| 238 |
+
00:03:36,965 --> 00:03:42,369
|
| 239 |
+
The 360 retained checkpoints do not cover every method or imply
|
| 240 |
+
|
| 241 |
+
61
|
| 242 |
+
00:03:42,369 --> 00:03:44,312
|
| 243 |
+
complete inference reruns.
|
| 244 |
+
|
| 245 |
+
62
|
| 246 |
+
00:03:44,312 --> 00:03:45,496
|
| 247 |
Explore the public releases.
|
videos/WISP_en.vtt
CHANGED
|
@@ -1,178 +1,187 @@
|
|
| 1 |
WEBVTT
|
| 2 |
|
| 3 |
-
00:00:
|
| 4 |
-
Walking, cooking, and handling objects
|
| 5 |
|
| 6 |
-
00:00:
|
| 7 |
-
|
| 8 |
|
| 9 |
-
00:00:
|
| 10 |
-
Local shape, rhythm, temporal order, and
|
| 11 |
|
| 12 |
-
00:00:
|
| 13 |
-
|
| 14 |
|
| 15 |
-
00:00:
|
| 16 |
-
When benchmarks mix sensor locations, they
|
| 17 |
|
| 18 |
-
00:00:
|
| 19 |
-
|
| 20 |
|
| 21 |
-
00:00:
|
| 22 |
-
WristHARBench brings together nine wrist
|
| 23 |
|
| 24 |
-
00:00:
|
| 25 |
-
|
| 26 |
|
| 27 |
-
00:00:
|
| 28 |
-
Two additional multichannel inertial
|
| 29 |
|
| 30 |
-
00:00:
|
| 31 |
-
|
| 32 |
|
| 33 |
-
00:00:
|
| 34 |
-
Within each repeat, different people
|
| 35 |
|
| 36 |
-
00:00:
|
| 37 |
-
|
| 38 |
|
| 39 |
-
00:00:
|
| 40 |
-
No person appears in two groups.
|
| 41 |
|
| 42 |
-
00:00:
|
| 43 |
-
|
| 44 |
|
| 45 |
-
00:00:
|
| 46 |
-
Methods share the same windows and splits,
|
| 47 |
|
| 48 |
-
00:00:
|
|
|
|
|
|
|
|
|
|
| 49 |
WISP reads the same signal window through complementary representations.
|
| 50 |
|
| 51 |
-
00:00:
|
| 52 |
-
Statistics and spectra describe overall variation and rhythm;
|
| 53 |
|
| 54 |
-
00:
|
| 55 |
-
|
| 56 |
|
| 57 |
-
00:
|
| 58 |
These landscapes illustrate operators on synthetic signals.
|
| 59 |
|
| 60 |
-
00:01:
|
| 61 |
-
Random convolution kernels move through
|
| 62 |
|
| 63 |
-
00:01:
|
| 64 |
-
|
| 65 |
|
| 66 |
-
00:01:
|
| 67 |
A selected slice reveals the responses within one signal window.
|
| 68 |
|
| 69 |
-
00:01:
|
| 70 |
-
Positive proportion, maximum, mean, and standard
|
| 71 |
|
| 72 |
-
00:01:
|
| 73 |
-
|
| 74 |
|
| 75 |
-
00:01:
|
| 76 |
-
The state-and-order representation maps
|
| 77 |
|
| 78 |
-
00:01:
|
| 79 |
-
|
| 80 |
|
| 81 |
-
00:01:
|
| 82 |
It counts how often states occur and how they transition to each other.
|
| 83 |
|
| 84 |
-
00:01:
|
| 85 |
The matrix contains transition probabilities, not attention weights.
|
| 86 |
|
| 87 |
-
00:01:
|
| 88 |
-
Geometry, interval, and shape features combine
|
| 89 |
|
| 90 |
-
00:01:
|
| 91 |
-
|
| 92 |
|
| 93 |
-
00:01:
|
| 94 |
-
A template moves along the window. The
|
| 95 |
|
| 96 |
-
00:01:
|
| 97 |
-
|
| 98 |
|
| 99 |
-
00:
|
| 100 |
Both match quality and location become features.
|
| 101 |
|
| 102 |
-
00:
|
| 103 |
Complementary clues can be combined at the feature level.
|
| 104 |
|
| 105 |
-
00:
|
| 106 |
-
|
|
|
|
|
|
|
|
|
|
| 107 |
|
| 108 |
-
00:
|
| 109 |
-
|
| 110 |
|
| 111 |
-
00:
|
| 112 |
-
|
| 113 |
|
| 114 |
-
00:02:
|
| 115 |
-
Score ensembles combine independently fitted
|
| 116 |
|
| 117 |
-
00:02:
|
| 118 |
-
|
| 119 |
|
| 120 |
-
00:02:
|
| 121 |
CSE averages three members equally; ESE uses weighted averaging.
|
| 122 |
|
| 123 |
-
00:02:
|
| 124 |
This is different from concatenating features before one classifier.
|
| 125 |
|
| 126 |
-
00:02:
|
| 127 |
-
On the frozen paper evaluation, WISP with
|
|
|
|
|
|
|
|
|
|
| 128 |
|
| 129 |
-
00:02:
|
| 130 |
-
|
| 131 |
|
| 132 |
-
00:02:
|
| 133 |
-
|
| 134 |
|
| 135 |
-
00:02:
|
| 136 |
-
|
| 137 |
|
| 138 |
-
00:
|
| 139 |
-
|
| 140 |
|
| 141 |
-
00:
|
| 142 |
-
|
| 143 |
|
| 144 |
-
00:
|
| 145 |
-
|
| 146 |
|
| 147 |
-
00:
|
| 148 |
-
With the other two groups already present,
|
| 149 |
|
| 150 |
-
00:
|
| 151 |
-
|
| 152 |
|
| 153 |
-
00:
|
| 154 |
-
Adding I changes the mean by minus 0.27
|
| 155 |
|
| 156 |
-
00:
|
| 157 |
-
|
| 158 |
|
| 159 |
-
00:
|
| 160 |
-
The evidence supports adding missing clues,
|
| 161 |
|
| 162 |
-
00:
|
| 163 |
-
|
| 164 |
|
| 165 |
-
00:03:
|
| 166 |
-
WristHARBench provides benchmark data. WISP provides
|
| 167 |
|
| 168 |
-
00:03:
|
| 169 |
-
|
| 170 |
|
| 171 |
-
00:03:
|
| 172 |
-
The 360 retained checkpoints do not cover
|
| 173 |
|
| 174 |
-
00:03:
|
| 175 |
-
|
| 176 |
|
| 177 |
-
00:03:
|
| 178 |
Explore the public releases.
|
|
|
|
| 1 |
WEBVTT
|
| 2 |
|
| 3 |
+
00:00:01.004 --> 00:00:04.350
|
| 4 |
+
Walking, cooking, and handling objects leave
|
| 5 |
|
| 6 |
+
00:00:04.350 --> 00:00:06.715
|
| 7 |
+
different clues in wrist motion.
|
| 8 |
|
| 9 |
+
00:00:06.715 --> 00:00:12.042
|
| 10 |
+
Local shape, rhythm, temporal order, and relationships between axes can
|
| 11 |
|
| 12 |
+
00:00:12.042 --> 00:00:13.812
|
| 13 |
+
all help identify an activity.
|
| 14 |
|
| 15 |
+
00:00:15.168 --> 00:00:18.956
|
| 16 |
+
When benchmarks mix sensor locations, they cannot isolate
|
| 17 |
|
| 18 |
+
00:00:18.956 --> 00:00:21.022
|
| 19 |
+
what the wrist alone reveals.
|
| 20 |
|
| 21 |
+
00:00:21.022 --> 00:00:25.264
|
| 22 |
+
WristHARBench brings together nine wrist acceleration datasets
|
| 23 |
|
| 24 |
+
00:00:25.264 --> 00:00:27.841
|
| 25 |
+
under one evaluation contract.
|
| 26 |
|
| 27 |
+
00:00:27.841 --> 00:00:31.187
|
| 28 |
+
Two additional multichannel inertial datasets support a
|
| 29 |
|
| 30 |
+
00:00:31.187 --> 00:00:32.398
|
| 31 |
+
separate sensitivity analysis.
|
| 32 |
|
| 33 |
+
00:00:34.064 --> 00:00:37.718
|
| 34 |
+
Within each repeat, different people enter training,
|
| 35 |
|
| 36 |
+
00:00:37.718 --> 00:00:39.929
|
| 37 |
+
validation, and test groups.
|
| 38 |
|
| 39 |
+
00:00:39.929 --> 00:00:43.814
|
| 40 |
+
No person appears in two groups. Validation guides model
|
| 41 |
|
| 42 |
+
00:00:43.814 --> 00:00:46.679
|
| 43 |
+
choice; testing comes last.
|
| 44 |
|
| 45 |
+
00:00:46.679 --> 00:00:49.198
|
| 46 |
+
Methods share the same windows and splits,
|
| 47 |
|
| 48 |
+
00:00:49.198 --> 00:00:51.045
|
| 49 |
+
across five repeated holdouts.
|
| 50 |
+
|
| 51 |
+
00:00:52.296 --> 00:00:57.815
|
| 52 |
WISP reads the same signal window through complementary representations.
|
| 53 |
|
| 54 |
+
00:00:57.815 --> 00:01:02.412
|
| 55 |
+
Statistics and spectra describe overall variation and rhythm; other
|
| 56 |
|
| 57 |
+
00:01:02.412 --> 00:01:06.527
|
| 58 |
+
branches retain shape, order, and cross-axis structure.
|
| 59 |
|
| 60 |
+
00:01:06.527 --> 00:01:09.642
|
| 61 |
These landscapes illustrate operators on synthetic signals.
|
| 62 |
|
| 63 |
+
00:01:11.141 --> 00:01:14.487
|
| 64 |
+
Random convolution kernels move through time, responding
|
| 65 |
|
| 66 |
+
00:01:14.487 --> 00:01:16.679
|
| 67 |
+
to local waveform patterns.
|
| 68 |
|
| 69 |
+
00:01:16.679 --> 00:01:21.391
|
| 70 |
A selected slice reveals the responses within one signal window.
|
| 71 |
|
| 72 |
+
00:01:21.391 --> 00:01:26.737
|
| 73 |
+
Positive proportion, maximum, mean, and standard deviation pool that
|
| 74 |
|
| 75 |
+
00:01:26.737 --> 00:01:28.301
|
| 76 |
+
response into four features.
|
| 77 |
|
| 78 |
+
00:01:29.505 --> 00:01:33.775
|
| 79 |
+
The state-and-order representation maps values to relative,
|
| 80 |
|
| 81 |
+
00:01:33.775 --> 00:01:35.775
|
| 82 |
+
within-window quantile states.
|
| 83 |
|
| 84 |
+
00:01:35.775 --> 00:01:40.025
|
| 85 |
It counts how often states occur and how they transition to each other.
|
| 86 |
|
| 87 |
+
00:01:40.025 --> 00:01:44.026
|
| 88 |
The matrix contains transition probabilities, not attention weights.
|
| 89 |
|
| 90 |
+
00:01:45.527 --> 00:01:51.546
|
| 91 |
+
Geometry, interval, and shape features combine cross-axis structure,
|
| 92 |
|
| 93 |
+
00:01:51.546 --> 00:01:54.969
|
| 94 |
+
local statistics, and subsequence matches.
|
| 95 |
|
| 96 |
+
00:01:54.969 --> 00:01:58.854
|
| 97 |
+
A template moves along the window. The lowest distance
|
| 98 |
|
| 99 |
+
00:01:58.854 --> 00:02:00.835
|
| 100 |
+
identifies its best match.
|
| 101 |
|
| 102 |
+
00:02:00.835 --> 00:02:04.058
|
| 103 |
Both match quality and location become features.
|
| 104 |
|
| 105 |
+
00:02:05.271 --> 00:02:09.214
|
| 106 |
Complementary clues can be combined at the feature level.
|
| 107 |
|
| 108 |
+
00:02:09.214 --> 00:02:15.060
|
| 109 |
+
CIS concatenates convolution features R, interval features I, and
|
| 110 |
+
|
| 111 |
+
00:02:15.060 --> 00:02:18.771
|
| 112 |
+
spectral, temporal, and cross-axis features S.
|
| 113 |
|
| 114 |
+
00:02:18.771 --> 00:02:21.368
|
| 115 |
+
A single classifier reads the combined vector
|
| 116 |
|
| 117 |
+
00:02:21.368 --> 00:02:23.156
|
| 118 |
+
and produces activity scores.
|
| 119 |
|
| 120 |
+
00:02:24.333 --> 00:02:27.679
|
| 121 |
+
Score ensembles combine independently fitted members after
|
| 122 |
|
| 123 |
+
00:02:27.679 --> 00:02:29.775
|
| 124 |
+
aligning their activity classes.
|
| 125 |
|
| 126 |
+
00:02:29.775 --> 00:02:35.910
|
| 127 |
CSE averages three members equally; ESE uses weighted averaging.
|
| 128 |
|
| 129 |
+
00:02:35.910 --> 00:02:39.371
|
| 130 |
This is different from concatenating features before one classifier.
|
| 131 |
|
| 132 |
+
00:02:40.719 --> 00:02:45.508
|
| 133 |
+
On the frozen paper evaluation, WISP with random search reaches
|
| 134 |
+
|
| 135 |
+
00:02:45.508 --> 00:02:49.642
|
| 136 |
+
70.20 percent mean macro F1.
|
| 137 |
|
| 138 |
+
00:02:49.642 --> 00:02:53.412
|
| 139 |
+
RDST, the strongest of eighteen external baselines on this
|
| 140 |
|
| 141 |
+
00:02:53.412 --> 00:02:57.027
|
| 142 |
+
average, reaches 59.62 percent.
|
| 143 |
|
| 144 |
+
00:02:57.027 --> 00:03:02.565
|
| 145 |
+
The difference is 10.58 percentage points. This is a dataset average,
|
| 146 |
|
| 147 |
+
00:03:02.565 --> 00:03:04.331
|
| 148 |
+
not a win on every dataset.
|
| 149 |
|
| 150 |
+
00:03:05.750 --> 00:03:07.962
|
| 151 |
+
Controlled ablations test complementary
|
| 152 |
|
| 153 |
+
00:03:07.962 --> 00:03:10.115
|
| 154 |
+
information with a fixed classifier.
|
| 155 |
|
| 156 |
+
00:03:10.115 --> 00:03:14.481
|
| 157 |
+
With the other two groups already present, adding R gains 4.80
|
| 158 |
|
| 159 |
+
00:03:14.481 --> 00:03:17.962
|
| 160 |
+
points; adding S gains 3.01.
|
| 161 |
|
| 162 |
+
00:03:17.962 --> 00:03:22.096
|
| 163 |
+
Adding I changes the mean by minus 0.27 points, with
|
| 164 |
|
| 165 |
+
00:03:22.096 --> 00:03:24.000
|
| 166 |
+
an interval crossing zero.
|
| 167 |
|
| 168 |
+
00:03:24.000 --> 00:03:27.308
|
| 169 |
+
The evidence supports adding missing clues, rather than more
|
| 170 |
|
| 171 |
+
00:03:27.308 --> 00:03:28.173
|
| 172 |
+
overlapping representations.
|
| 173 |
|
| 174 |
+
00:03:29.658 --> 00:03:34.523
|
| 175 |
+
WristHARBench provides benchmark data. WISP provides method code, saved
|
| 176 |
|
| 177 |
+
00:03:34.523 --> 00:03:36.965
|
| 178 |
+
results, and retained search models.
|
| 179 |
|
| 180 |
+
00:03:36.965 --> 00:03:42.369
|
| 181 |
+
The 360 retained checkpoints do not cover every method or imply
|
| 182 |
|
| 183 |
+
00:03:42.369 --> 00:03:44.312
|
| 184 |
+
complete inference reruns.
|
| 185 |
|
| 186 |
+
00:03:44.312 --> 00:03:45.496
|
| 187 |
Explore the public releases.
|
videos/WISP_en_mimo_voice.mp4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a84479dd6c90c65300f8a681c7dcf3a9cc2ef4cb023512f3ebf3924f3596cb1
|
| 3 |
+
size 102486414
|
videos/WISP_en_mimo_voice.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5707f14bbc83074c5aeea409c88faa1faed9ad6d7c804cd8857d7150c1e80081
|
| 3 |
+
size 21702444
|
videos/WISP_en_script.txt
CHANGED
|
@@ -1,82 +1,59 @@
|
|
| 1 |
00:00:00,000 Reading activity at the wrist
|
| 2 |
-
Walking, cooking, and handling objects
|
| 3 |
-
|
| 4 |
-
Local shape, rhythm, temporal order, and
|
| 5 |
-
relationships between axes can all help identify an activity.
|
| 6 |
|
| 7 |
-
00:00:
|
| 8 |
-
When benchmarks mix sensor locations, they
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
acceleration datasets under one evaluation contract.
|
| 12 |
-
Two additional multichannel inertial
|
| 13 |
-
datasets support a separate sensitivity analysis.
|
| 14 |
|
| 15 |
-
00:00:
|
| 16 |
-
Within each repeat, different people
|
| 17 |
-
|
| 18 |
-
No person appears in two groups.
|
| 19 |
-
Validation guides model choice; testing comes last.
|
| 20 |
Methods share the same windows and splits, across five repeated holdouts.
|
| 21 |
|
| 22 |
-
00:00:
|
| 23 |
WISP reads the same signal window through complementary representations.
|
| 24 |
-
Statistics and spectra describe overall variation and rhythm;
|
| 25 |
-
other branches retain shape, order, and cross-axis structure.
|
| 26 |
These landscapes illustrate operators on synthetic signals.
|
| 27 |
|
| 28 |
-
00:01:
|
| 29 |
-
Random convolution kernels move through
|
| 30 |
-
time, responding to local waveform patterns.
|
| 31 |
A selected slice reveals the responses within one signal window.
|
| 32 |
-
Positive proportion, maximum, mean, and standard
|
| 33 |
-
deviation pool that response into four features.
|
| 34 |
|
| 35 |
-
00:01:
|
| 36 |
-
The state-and-order representation maps
|
| 37 |
-
values to relative, within-window quantile states.
|
| 38 |
It counts how often states occur and how they transition to each other.
|
| 39 |
The matrix contains transition probabilities, not attention weights.
|
| 40 |
|
| 41 |
-
00:01:
|
| 42 |
-
Geometry, interval, and shape features combine
|
| 43 |
-
|
| 44 |
-
A template moves along the window. The
|
| 45 |
-
lowest distance identifies its best match.
|
| 46 |
Both match quality and location become features.
|
| 47 |
|
| 48 |
-
00:
|
| 49 |
Complementary clues can be combined at the feature level.
|
| 50 |
-
features I, and spectral, temporal, and cross-axis features S.
|
| 51 |
-
features I, and spectral, temporal, and cross-axis
|
| 52 |
A single classifier reads the combined vector and produces activity scores.
|
| 53 |
|
| 54 |
-
00:02:
|
| 55 |
-
Score ensembles combine independently fitted
|
| 56 |
-
members after aligning their activity classes.
|
| 57 |
CSE averages three members equally; ESE uses weighted averaging.
|
| 58 |
This is different from concatenating features before one classifier.
|
| 59 |
|
| 60 |
-
00:02:
|
| 61 |
-
On the frozen paper evaluation, WISP with
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
baselines on this average, reaches 59.62 percent.
|
| 65 |
-
The difference is 10.58 percentage points. This
|
| 66 |
-
is a dataset average, not a win on every dataset.
|
| 67 |
|
| 68 |
-
00:
|
| 69 |
Controlled ablations test complementary information with a fixed classifier.
|
| 70 |
-
With the other two groups already present,
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
WristHARBench provides benchmark data. WISP provides
|
| 79 |
-
method code, saved results, and retained search models.
|
| 80 |
-
The 360 retained checkpoints do not cover
|
| 81 |
-
every method or imply complete inference reruns.
|
| 82 |
-
Explore the public releases.
|
|
|
|
| 1 |
00:00:00,000 Reading activity at the wrist
|
| 2 |
+
Walking, cooking, and handling objects leave different clues in wrist motion.
|
| 3 |
+
Local shape, rhythm, temporal order, and relationships between axes can all help identify an activity.
|
|
|
|
|
|
|
| 4 |
|
| 5 |
+
00:00:14,433 WristHARBench: a common comparison
|
| 6 |
+
When benchmarks mix sensor locations, they cannot isolate what the wrist alone reveals.
|
| 7 |
+
WristHARBench brings together nine wrist acceleration datasets under one evaluation contract.
|
| 8 |
+
Two additional multichannel inertial datasets support a separate sensitivity analysis.
|
|
|
|
|
|
|
|
|
|
| 9 |
|
| 10 |
+
00:00:33,233 Separate people before comparing models
|
| 11 |
+
Within each repeat, different people enter training, validation, and test groups.
|
| 12 |
+
No person appears in two groups. Validation guides model choice; testing comes last.
|
|
|
|
|
|
|
| 13 |
Methods share the same windows and splits, across five repeated holdouts.
|
| 14 |
|
| 15 |
+
00:00:51,600 One window, several readings
|
| 16 |
WISP reads the same signal window through complementary representations.
|
| 17 |
+
Statistics and spectra describe overall variation and rhythm; other branches retain shape, order, and cross-axis structure.
|
|
|
|
| 18 |
These landscapes illustrate operators on synthetic signals.
|
| 19 |
|
| 20 |
+
00:01:10,233 RC: from responses to features
|
| 21 |
+
Random convolution kernels move through time, responding to local waveform patterns.
|
|
|
|
| 22 |
A selected slice reveals the responses within one signal window.
|
| 23 |
+
Positive proportion, maximum, mean, and standard deviation pool that response into four features.
|
|
|
|
| 24 |
|
| 25 |
+
00:01:28,867 SO: relative states and their order
|
| 26 |
+
The state-and-order representation maps values to relative, within-window quantile states.
|
|
|
|
| 27 |
It counts how often states occur and how they transition to each other.
|
| 28 |
The matrix contains transition probabilities, not attention weights.
|
| 29 |
|
| 30 |
+
00:01:44,600 GIS: finding shapes within a window
|
| 31 |
+
Geometry, interval, and shape features combine cross-axis structure, local statistics, and subsequence matches.
|
| 32 |
+
A template moves along the window. The lowest distance identifies its best match.
|
|
|
|
|
|
|
| 33 |
Both match quality and location become features.
|
| 34 |
|
| 35 |
+
00:02:04,633 CIS: concatenate features first
|
| 36 |
Complementary clues can be combined at the feature level.
|
| 37 |
+
CIS concatenates convolution features R, interval features I, and spectral, temporal, and cross-axis features S.
|
|
|
|
| 38 |
A single classifier reads the combined vector and produces activity scores.
|
| 39 |
|
| 40 |
+
00:02:23,733 CSE / ESE: combine class scores
|
| 41 |
+
Score ensembles combine independently fitted members after aligning their activity classes.
|
|
|
|
| 42 |
CSE averages three members equally; ESE uses weighted averaging.
|
| 43 |
This is different from concatenating features before one classifier.
|
| 44 |
|
| 45 |
+
00:02:40,100 Nine datasets, equal weight
|
| 46 |
+
On the frozen paper evaluation, WISP with random search reaches 70.20 percent mean macro F1.
|
| 47 |
+
RDST, the strongest of eighteen external baselines on this average, reaches 59.62 percent.
|
| 48 |
+
The difference is 10.58 percentage points. This is a dataset average, not a win on every dataset.
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
+
00:03:04,900 What information does a branch add?
|
| 51 |
Controlled ablations test complementary information with a fixed classifier.
|
| 52 |
+
With the other two groups already present, adding R gains 4.80 points; adding S gains 3.01.
|
| 53 |
+
Adding I changes the mean by minus 0.27 points, with an interval crossing zero.
|
| 54 |
+
The evidence supports adding missing clues, rather than more overlapping representations.
|
| 55 |
+
|
| 56 |
+
00:03:29,000 From shared evaluation to composable representations
|
| 57 |
+
WristHARBench provides benchmark data. WISP provides method code, saved results, and retained search models.
|
| 58 |
+
The 360 retained checkpoints do not cover every method or imply complete inference reruns.
|
| 59 |
+
Explore the public releases.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
videos/WISP_en_system_voice.mp4
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a84479dd6c90c65300f8a681c7dcf3a9cc2ef4cb023512f3ebf3924f3596cb1
|
| 3 |
+
size 102486414
|
videos/WISP_zh.srt
CHANGED
|
@@ -1,183 +1,187 @@
|
|
| 1 |
1
|
| 2 |
-
00:00:00,
|
| 3 |
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 4 |
|
| 5 |
2
|
| 6 |
-
00:00:
|
| 7 |
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 8 |
|
| 9 |
3
|
| 10 |
-
00:00:
|
| 11 |
都可能帮助识别活动。
|
| 12 |
|
| 13 |
4
|
| 14 |
-
00:00:
|
| 15 |
-
如果把身体不同部位的传感器混在一起,
|
| 16 |
|
| 17 |
5
|
| 18 |
-
00:00:
|
| 19 |
-
究竟能识别什么?
|
| 20 |
|
| 21 |
6
|
| 22 |
-
00:00:
|
| 23 |
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 24 |
|
| 25 |
7
|
| 26 |
-
00:00:
|
| 27 |
另有两个多通道惯性数据集,单独分析。
|
| 28 |
|
| 29 |
8
|
| 30 |
-
00:00:
|
| 31 |
每次划分中,不同参与者进入训练、验证和测试组,
|
| 32 |
|
| 33 |
9
|
| 34 |
-
00:00:
|
| 35 |
同一个人不会跨组出现。
|
| 36 |
|
| 37 |
10
|
| 38 |
-
00:00:
|
| 39 |
模型设置只用验证组选择,测试组留到最后。
|
| 40 |
|
| 41 |
11
|
| 42 |
-
00:00:
|
| 43 |
所有方法使用相同的窗口和划分,重复五次。
|
| 44 |
|
| 45 |
12
|
| 46 |
-
00:00:
|
| 47 |
WISP 把同一段信号转换为多种表征。
|
| 48 |
|
| 49 |
13
|
| 50 |
-
00:00:
|
| 51 |
统计和频谱描述整体变化与节律;
|
| 52 |
|
| 53 |
14
|
| 54 |
-
00:
|
| 55 |
其他分支保留局部形状、相对次序和轴间关系。
|
| 56 |
|
| 57 |
15
|
| 58 |
-
00:
|
| 59 |
这里的曲面来自合成信号,是机制示意。
|
| 60 |
|
| 61 |
16
|
| 62 |
-
00:01:
|
| 63 |
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 64 |
|
| 65 |
17
|
| 66 |
-
00:01:
|
| 67 |
选定一条截线,就能看到同一窗口内的响应变化。
|
| 68 |
|
| 69 |
18
|
| 70 |
-
00:01:
|
| 71 |
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 72 |
|
| 73 |
19
|
| 74 |
-
00:01:
|
| 75 |
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 76 |
|
| 77 |
20
|
| 78 |
-
00:01:
|
| 79 |
再统计各状态出现多少,以及它们如何相互转移。
|
| 80 |
|
| 81 |
21
|
| 82 |
-
00:01:
|
| 83 |
矩阵记录的是转移概率,不是注意力权重。
|
| 84 |
|
| 85 |
22
|
| 86 |
-
00:01:
|
| 87 |
-
几何、区间和形状表征,
|
| 88 |
|
| 89 |
23
|
| 90 |
-
00:01:
|
| 91 |
-
|
| 92 |
|
| 93 |
24
|
| 94 |
-
00:01:
|
| 95 |
-
|
| 96 |
|
| 97 |
25
|
| 98 |
-
00:
|
| 99 |
-
|
| 100 |
|
| 101 |
26
|
| 102 |
-
00:
|
| 103 |
-
|
| 104 |
|
| 105 |
27
|
| 106 |
-
00:
|
| 107 |
-
|
| 108 |
|
| 109 |
28
|
| 110 |
-
00:
|
| 111 |
-
|
| 112 |
|
| 113 |
29
|
| 114 |
-
00:02:
|
| 115 |
-
|
| 116 |
|
| 117 |
30
|
| 118 |
-
00:02:
|
| 119 |
-
|
| 120 |
|
| 121 |
31
|
| 122 |
-
00:02:
|
| 123 |
-
|
| 124 |
|
| 125 |
32
|
| 126 |
-
00:02:
|
| 127 |
-
|
| 128 |
|
| 129 |
33
|
| 130 |
-
00:02:
|
| 131 |
-
|
| 132 |
|
| 133 |
34
|
| 134 |
-
00:02:
|
| 135 |
-
|
| 136 |
|
| 137 |
35
|
| 138 |
-
00:02:
|
| 139 |
-
|
| 140 |
|
| 141 |
36
|
| 142 |
-
00:02:
|
| 143 |
-
|
| 144 |
|
| 145 |
37
|
| 146 |
-
00:02:
|
| 147 |
-
|
| 148 |
|
| 149 |
38
|
| 150 |
-
00:
|
| 151 |
-
|
| 152 |
|
| 153 |
39
|
| 154 |
-
00:
|
| 155 |
-
|
| 156 |
|
| 157 |
40
|
| 158 |
-
00:
|
| 159 |
-
|
| 160 |
|
| 161 |
41
|
| 162 |
-
00:
|
| 163 |
-
|
| 164 |
|
| 165 |
42
|
| 166 |
-
00:
|
| 167 |
-
|
| 168 |
|
| 169 |
43
|
| 170 |
-
00:
|
| 171 |
-
|
| 172 |
|
| 173 |
44
|
| 174 |
-
00:03:
|
| 175 |
-
|
| 176 |
|
| 177 |
45
|
| 178 |
-
00:03:
|
| 179 |
-
|
| 180 |
|
| 181 |
46
|
| 182 |
-
00:03:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 183 |
从公开入口,继续探索。
|
|
|
|
| 1 |
1
|
| 2 |
+
00:00:00,650 --> 00:00:07,868
|
| 3 |
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 4 |
|
| 5 |
2
|
| 6 |
+
00:00:07,868 --> 00:00:12,396
|
| 7 |
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 8 |
|
| 9 |
3
|
| 10 |
+
00:00:12,396 --> 00:00:13,796
|
| 11 |
都可能帮助识别活动。
|
| 12 |
|
| 13 |
4
|
| 14 |
+
00:00:15,015 --> 00:00:18,725
|
| 15 |
+
如果把身体不同部位的传感器混在一起,
|
| 16 |
|
| 17 |
5
|
| 18 |
+
00:00:18,725 --> 00:00:23,541
|
| 19 |
+
结果就难以回答:只看手腕,究竟能识别什么?
|
| 20 |
|
| 21 |
6
|
| 22 |
+
00:00:23,541 --> 00:00:28,543
|
| 23 |
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 24 |
|
| 25 |
7
|
| 26 |
+
00:00:28,543 --> 00:00:31,560
|
| 27 |
另有两个多通道惯性数据集,单独分析。
|
| 28 |
|
| 29 |
8
|
| 30 |
+
00:00:32,721 --> 00:00:38,231
|
| 31 |
每次划分中,不同参与者进入训练、验证和测试组,
|
| 32 |
|
| 33 |
9
|
| 34 |
+
00:00:38,231 --> 00:00:40,885
|
| 35 |
同一个人不会跨组出现。
|
| 36 |
|
| 37 |
10
|
| 38 |
+
00:00:40,885 --> 00:00:45,776
|
| 39 |
模型设置只用验证组选择,测试组留到最后。
|
| 40 |
|
| 41 |
11
|
| 42 |
+
00:00:45,776 --> 00:00:49,406
|
| 43 |
所有方法使用相同的窗口和划分,重复五次。
|
| 44 |
|
| 45 |
12
|
| 46 |
+
00:00:50,709 --> 00:00:56,600
|
| 47 |
WISP 把同一段信号转换为多种表征。
|
| 48 |
|
| 49 |
13
|
| 50 |
+
00:00:56,600 --> 00:01:00,837
|
| 51 |
统计和频谱描述整体变化与节律;
|
| 52 |
|
| 53 |
14
|
| 54 |
+
00:01:00,837 --> 00:01:06,764
|
| 55 |
其他分支保留局部形状、相对次序和轴间关系。
|
| 56 |
|
| 57 |
15
|
| 58 |
+
00:01:06,764 --> 00:01:10,657
|
| 59 |
这里的曲面来自合成信号,是机制示意。
|
| 60 |
|
| 61 |
16
|
| 62 |
+
00:01:11,865 --> 00:01:18,138
|
| 63 |
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 64 |
|
| 65 |
17
|
| 66 |
+
00:01:18,138 --> 00:01:23,120
|
| 67 |
选定一条截线,就能看到同一窗口内的响应变化。
|
| 68 |
|
| 69 |
18
|
| 70 |
+
00:01:23,120 --> 00:01:29,938
|
| 71 |
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 72 |
|
| 73 |
19
|
| 74 |
+
00:01:31,146 --> 00:01:38,964
|
| 75 |
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 76 |
|
| 77 |
20
|
| 78 |
+
00:01:38,964 --> 00:01:44,046
|
| 79 |
再统计各状态出现多少,以及它们如何相互转移。
|
| 80 |
|
| 81 |
21
|
| 82 |
+
00:01:44,046 --> 00:01:48,917
|
| 83 |
矩阵记录的是转移概率,不是注意力权重。
|
| 84 |
|
| 85 |
22
|
| 86 |
+
00:01:50,104 --> 00:01:53,231
|
| 87 |
+
几何、区间和形状表征,
|
| 88 |
|
| 89 |
23
|
| 90 |
+
00:01:53,231 --> 00:01:58,849
|
| 91 |
+
同时读取轴间结构、局部统计和子序列匹配。
|
| 92 |
|
| 93 |
24
|
| 94 |
+
00:01:58,849 --> 00:02:03,940
|
| 95 |
+
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 96 |
|
| 97 |
25
|
| 98 |
+
00:02:03,940 --> 00:02:07,558
|
| 99 |
+
保留下来的线索,包括匹配质量和出现位置。
|
| 100 |
|
| 101 |
26
|
| 102 |
+
00:02:08,735 --> 00:02:12,590
|
| 103 |
+
互补线索可以在特征层组合。
|
| 104 |
|
| 105 |
27
|
| 106 |
+
00:02:12,590 --> 00:02:16,844
|
| 107 |
+
CIS 把卷积特征 R、区间特征 I,
|
| 108 |
|
| 109 |
28
|
| 110 |
+
00:02:16,844 --> 00:02:21,826
|
| 111 |
+
以及频谱、时间和轴间特征 S,拼接成一个向量。
|
| 112 |
|
| 113 |
29
|
| 114 |
+
00:02:21,826 --> 00:02:25,572
|
| 115 |
+
随后由同一个分类器读取,输出各活动的分数。
|
| 116 |
|
| 117 |
30
|
| 118 |
+
00:02:26,918 --> 00:02:31,791
|
| 119 |
+
另一种组合发生在分数层:各成员先独立预测,
|
| 120 |
|
| 121 |
31
|
| 122 |
+
00:02:31,791 --> 00:02:34,882
|
| 123 |
+
再按相同的活动顺序对齐。
|
| 124 |
|
| 125 |
32
|
| 126 |
+
00:02:34,882 --> 00:02:40,046
|
| 127 |
+
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 128 |
|
| 129 |
33
|
| 130 |
+
00:02:40,046 --> 00:02:42,936
|
| 131 |
+
这与特征拼接,是两种不同的组合方式。
|
| 132 |
|
| 133 |
34
|
| 134 |
+
00:02:44,409 --> 00:02:46,864
|
| 135 |
+
在冻结的论文评估范围内,
|
| 136 |
|
| 137 |
35
|
| 138 |
+
00:02:46,864 --> 00:02:53,421
|
| 139 |
+
随机搜索版 WISP 的平均 Macro F1 为百分之七十点二零。
|
| 140 |
|
| 141 |
36
|
| 142 |
+
00:02:53,421 --> 00:02:58,573
|
| 143 |
+
十八个外部基线中,最高的 RDST 为百分之五十九点六二,
|
| 144 |
|
| 145 |
37
|
| 146 |
+
00:02:58,573 --> 00:03:01,955
|
| 147 |
+
相差十点五八个百分点。
|
| 148 |
|
| 149 |
38
|
| 150 |
+
00:03:01,955 --> 00:03:06,155
|
| 151 |
+
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 152 |
|
| 153 |
39
|
| 154 |
+
00:03:07,419 --> 00:03:11,110
|
| 155 |
+
受控消融进一步检验互补性。
|
| 156 |
|
| 157 |
40
|
| 158 |
+
00:03:11,110 --> 00:03:15,256
|
| 159 |
+
当另外两组特征已经存在时,加入 R 和 S,
|
| 160 |
|
| 161 |
41
|
| 162 |
+
00:03:15,256 --> 00:03:20,419
|
| 163 |
+
分别平均增加四点八零和三点零一个百分点。
|
| 164 |
|
| 165 |
42
|
| 166 |
+
00:03:20,419 --> 00:03:25,801
|
| 167 |
+
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 168 |
|
| 169 |
43
|
| 170 |
+
00:03:25,801 --> 00:03:30,365
|
| 171 |
+
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 172 |
|
| 173 |
44
|
| 174 |
+
00:03:31,505 --> 00:03:34,396
|
| 175 |
+
WristHARBench 提供基准数据;
|
| 176 |
|
| 177 |
45
|
| 178 |
+
00:03:34,396 --> 00:03:39,719
|
| 179 |
+
WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
| 180 |
|
| 181 |
46
|
| 182 |
+
00:03:39,719 --> 00:03:44,778
|
| 183 |
+
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 184 |
+
|
| 185 |
+
47
|
| 186 |
+
00:03:44,778 --> 00:03:46,542
|
| 187 |
从公开入口,继续探索。
|
videos/WISP_zh.vtt
CHANGED
|
@@ -1,139 +1,142 @@
|
|
| 1 |
WEBVTT
|
| 2 |
|
| 3 |
-
00:00:00.
|
| 4 |
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 5 |
|
| 6 |
-
00:00:
|
| 7 |
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 8 |
|
| 9 |
-
00:00:
|
| 10 |
都可能帮助识别活动。
|
| 11 |
|
| 12 |
-
00:00:
|
| 13 |
-
如果把身体不同部位的传感器混在一起,
|
| 14 |
|
| 15 |
-
00:00:
|
| 16 |
-
究竟能识别什么?
|
| 17 |
|
| 18 |
-
00:00:
|
| 19 |
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 20 |
|
| 21 |
-
00:00:
|
| 22 |
另有两个多通道惯性数据集,单独分析。
|
| 23 |
|
| 24 |
-
00:00:
|
| 25 |
每次划分中,不同参与者进入训练、验证和测试组,
|
| 26 |
|
| 27 |
-
00:00:
|
| 28 |
同一个人不会跨组出现。
|
| 29 |
|
| 30 |
-
00:00:
|
| 31 |
模型设置只用验证组选择,测试组留到最后。
|
| 32 |
|
| 33 |
-
00:00:
|
| 34 |
所有方法使用相同的窗口和划分,重复五次。
|
| 35 |
|
| 36 |
-
00:00:
|
| 37 |
WISP 把同一段信号转换为多种表征。
|
| 38 |
|
| 39 |
-
00:00:
|
| 40 |
统计和频谱描述整体变化与节律;
|
| 41 |
|
| 42 |
-
00:
|
| 43 |
其他分支保留局部形状、相对次序和轴间关系。
|
| 44 |
|
| 45 |
-
00:
|
| 46 |
这里的曲面来自合成信号,是机制示意。
|
| 47 |
|
| 48 |
-
00:01:
|
| 49 |
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 50 |
|
| 51 |
-
00:01:
|
| 52 |
选定一条截线,就能看到同一窗口内的响应变化。
|
| 53 |
|
| 54 |
-
00:01:
|
| 55 |
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 56 |
|
| 57 |
-
00:01:
|
| 58 |
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 59 |
|
| 60 |
-
00:01:
|
| 61 |
再统计各状态出现多少,以及它们如何相互转移。
|
| 62 |
|
| 63 |
-
00:01:
|
| 64 |
矩阵记录的是转移概率,不是注意力权重。
|
| 65 |
|
| 66 |
-
00:01:
|
| 67 |
-
几何、区间和形状表征,
|
| 68 |
|
| 69 |
-
00:01:
|
|
|
|
|
|
|
|
|
|
| 70 |
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 71 |
|
| 72 |
-
00:
|
| 73 |
保留下来的线索,包括匹配质量和出现位置。
|
| 74 |
|
| 75 |
-
00:
|
| 76 |
互补线索可以在特征层组合。
|
| 77 |
|
| 78 |
-
00:
|
| 79 |
CIS 把卷积特征 R、区间特征 I,
|
| 80 |
|
| 81 |
-
00:
|
| 82 |
以及频谱、时间和轴间特征 S,拼接成一个向量。
|
| 83 |
|
| 84 |
-
00:
|
| 85 |
随后由同一个分类器读取,输出各活动的分数。
|
| 86 |
|
| 87 |
-
00:02:
|
| 88 |
另一种组合发生在分数层:各成员先独立预测,
|
| 89 |
|
| 90 |
-
00:02:
|
| 91 |
再按相同的活动顺序对齐。
|
| 92 |
|
| 93 |
-
00:02:
|
| 94 |
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 95 |
|
| 96 |
-
00:02:
|
| 97 |
这与特征拼接,是两种不同的组合方式。
|
| 98 |
|
| 99 |
-
00:02:
|
| 100 |
在冻结的论文评估范围内,
|
| 101 |
|
| 102 |
-
00:02:
|
| 103 |
随机搜索版 WISP 的平均 Macro F1 为百分之七十点二零。
|
| 104 |
|
| 105 |
-
00:02:
|
| 106 |
十八个外部基线中,最高的 RDST 为百分之五十九点六二,
|
| 107 |
|
| 108 |
-
00:02:
|
| 109 |
相差十点五八个百分点。
|
| 110 |
|
| 111 |
-
00:
|
| 112 |
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 113 |
|
| 114 |
-
00:
|
| 115 |
受控消融进一步检验互补性。
|
| 116 |
|
| 117 |
-
00:
|
| 118 |
当另外两组特征已���存在时,加入 R 和 S,
|
| 119 |
|
| 120 |
-
00:
|
| 121 |
分别平均增加四点八零和三点零一个百分点。
|
| 122 |
|
| 123 |
-
00:
|
| 124 |
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 125 |
|
| 126 |
-
00:
|
| 127 |
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 128 |
|
| 129 |
-
00:
|
| 130 |
WristHARBench 提供基准数据;
|
| 131 |
|
| 132 |
-
00:03:
|
| 133 |
WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
| 134 |
|
| 135 |
-
00:03:
|
| 136 |
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 137 |
|
| 138 |
-
00:03:
|
| 139 |
从公开入口,继续探索。
|
|
|
|
| 1 |
WEBVTT
|
| 2 |
|
| 3 |
+
00:00:00.650 --> 00:00:07.868
|
| 4 |
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 5 |
|
| 6 |
+
00:00:07.868 --> 00:00:12.396
|
| 7 |
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 8 |
|
| 9 |
+
00:00:12.396 --> 00:00:13.796
|
| 10 |
都可能帮助识别活动。
|
| 11 |
|
| 12 |
+
00:00:15.015 --> 00:00:18.725
|
| 13 |
+
如果把身体不同部位的传感器混在一起,
|
| 14 |
|
| 15 |
+
00:00:18.725 --> 00:00:23.541
|
| 16 |
+
结果就难以回答:只看手腕,究竟能识别什么?
|
| 17 |
|
| 18 |
+
00:00:23.541 --> 00:00:28.543
|
| 19 |
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 20 |
|
| 21 |
+
00:00:28.543 --> 00:00:31.560
|
| 22 |
另有两个多通道惯性数据集,单独分析。
|
| 23 |
|
| 24 |
+
00:00:32.721 --> 00:00:38.231
|
| 25 |
每次划分中,不同参与者进入训练、验证和测试组,
|
| 26 |
|
| 27 |
+
00:00:38.231 --> 00:00:40.885
|
| 28 |
同一个人不会跨组出现。
|
| 29 |
|
| 30 |
+
00:00:40.885 --> 00:00:45.776
|
| 31 |
模型设置只用验证组选择,测试组留到最后。
|
| 32 |
|
| 33 |
+
00:00:45.776 --> 00:00:49.406
|
| 34 |
所有方法使用相同的窗口和划分,重复五次。
|
| 35 |
|
| 36 |
+
00:00:50.709 --> 00:00:56.600
|
| 37 |
WISP 把同一段信号转换为多种表征。
|
| 38 |
|
| 39 |
+
00:00:56.600 --> 00:01:00.837
|
| 40 |
统计和频谱描述整体变化与节律;
|
| 41 |
|
| 42 |
+
00:01:00.837 --> 00:01:06.764
|
| 43 |
其他分支保留局部形状、相对次序和轴间关系。
|
| 44 |
|
| 45 |
+
00:01:06.764 --> 00:01:10.657
|
| 46 |
这里的曲面来自合成信号,是机制示意。
|
| 47 |
|
| 48 |
+
00:01:11.865 --> 00:01:18.138
|
| 49 |
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 50 |
|
| 51 |
+
00:01:18.138 --> 00:01:23.120
|
| 52 |
选定一条截线,就能看到同一窗口内的响应变化。
|
| 53 |
|
| 54 |
+
00:01:23.120 --> 00:01:29.938
|
| 55 |
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 56 |
|
| 57 |
+
00:01:31.146 --> 00:01:38.964
|
| 58 |
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 59 |
|
| 60 |
+
00:01:38.964 --> 00:01:44.046
|
| 61 |
再统计各状态出现多少,以及它们如何相互转移。
|
| 62 |
|
| 63 |
+
00:01:44.046 --> 00:01:48.917
|
| 64 |
矩阵记录的是转移概率,不是注意力权重。
|
| 65 |
|
| 66 |
+
00:01:50.104 --> 00:01:53.231
|
| 67 |
+
几何、区间和形状表征,
|
| 68 |
|
| 69 |
+
00:01:53.231 --> 00:01:58.849
|
| 70 |
+
同时读取轴间结构、局部统计和子序列匹配。
|
| 71 |
+
|
| 72 |
+
00:01:58.849 --> 00:02:03.940
|
| 73 |
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 74 |
|
| 75 |
+
00:02:03.940 --> 00:02:07.558
|
| 76 |
保留下来的线索,包括匹配质量和出现位置。
|
| 77 |
|
| 78 |
+
00:02:08.735 --> 00:02:12.590
|
| 79 |
互补线索可以在特征层组合。
|
| 80 |
|
| 81 |
+
00:02:12.590 --> 00:02:16.844
|
| 82 |
CIS 把卷积特征 R、区间特征 I,
|
| 83 |
|
| 84 |
+
00:02:16.844 --> 00:02:21.826
|
| 85 |
以及频谱、时间和轴间特征 S,拼接成一个向量。
|
| 86 |
|
| 87 |
+
00:02:21.826 --> 00:02:25.572
|
| 88 |
随后由同一个分类器读取,输出各活动的分数。
|
| 89 |
|
| 90 |
+
00:02:26.918 --> 00:02:31.791
|
| 91 |
另一种组合发生在分数层:各成员先独立预测,
|
| 92 |
|
| 93 |
+
00:02:31.791 --> 00:02:34.882
|
| 94 |
再按相同的活动顺序对齐。
|
| 95 |
|
| 96 |
+
00:02:34.882 --> 00:02:40.046
|
| 97 |
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 98 |
|
| 99 |
+
00:02:40.046 --> 00:02:42.936
|
| 100 |
这与特征拼接,是两种不同的组合方式。
|
| 101 |
|
| 102 |
+
00:02:44.409 --> 00:02:46.864
|
| 103 |
在冻结的论文评估范围内,
|
| 104 |
|
| 105 |
+
00:02:46.864 --> 00:02:53.421
|
| 106 |
随机搜索版 WISP 的平均 Macro F1 为百分之七十点二零。
|
| 107 |
|
| 108 |
+
00:02:53.421 --> 00:02:58.573
|
| 109 |
十八个外部基线中,最高的 RDST 为百分之五十九点六二,
|
| 110 |
|
| 111 |
+
00:02:58.573 --> 00:03:01.955
|
| 112 |
相差十点五八个百分点。
|
| 113 |
|
| 114 |
+
00:03:01.955 --> 00:03:06.155
|
| 115 |
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 116 |
|
| 117 |
+
00:03:07.419 --> 00:03:11.110
|
| 118 |
受控消融进一步检验互补性。
|
| 119 |
|
| 120 |
+
00:03:11.110 --> 00:03:15.256
|
| 121 |
当另外两组特征已���存在时,加入 R 和 S,
|
| 122 |
|
| 123 |
+
00:03:15.256 --> 00:03:20.419
|
| 124 |
分别平均增加四点八零和三点零一个百分点。
|
| 125 |
|
| 126 |
+
00:03:20.419 --> 00:03:25.801
|
| 127 |
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 128 |
|
| 129 |
+
00:03:25.801 --> 00:03:30.365
|
| 130 |
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 131 |
|
| 132 |
+
00:03:31.505 --> 00:03:34.396
|
| 133 |
WristHARBench 提供基准数据;
|
| 134 |
|
| 135 |
+
00:03:34.396 --> 00:03:39.719
|
| 136 |
WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
| 137 |
|
| 138 |
+
00:03:39.719 --> 00:03:44.778
|
| 139 |
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 140 |
|
| 141 |
+
00:03:44.778 --> 00:03:46.542
|
| 142 |
从公开入口,继续探索。
|
videos/WISP_zh_mimo_voice.mp4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a06a03c57cfc7e2cd6936fd232a7e4ee931255250eb08a4a16eff32635a1f8c1
|
| 3 |
+
size 105702723
|
videos/WISP_zh_mimo_voice.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b9ef2e4d01bc0527ecc2ac2edfc5377c0a261b05e8d39e08ed9fd6d8e4f5726
|
| 3 |
+
size 21804844
|
videos/WISP_zh_script.txt
CHANGED
|
@@ -1,69 +1,59 @@
|
|
| 1 |
00:00:00,000 从腕部信号,读懂活动
|
| 2 |
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 3 |
-
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 4 |
-
都可能帮助识别活动。
|
| 5 |
|
| 6 |
-
00:00:
|
| 7 |
-
如果把身体不同部位的传感器混在一起,结果就难以回答:只看手腕,
|
| 8 |
-
究竟能识别什么?
|
| 9 |
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 10 |
另有两个多通道惯性数据集,单独分析。
|
| 11 |
|
| 12 |
-
00:00:
|
| 13 |
-
每次划分中,不同参与者进入训练、验证和测试组,
|
| 14 |
-
同一个人不会跨组出现。
|
| 15 |
模型设置只用验证组选择,测试组留到最后。
|
| 16 |
所有方法使用相同的窗口和划分,重复五次。
|
| 17 |
|
| 18 |
-
00:00:
|
| 19 |
WISP 把同一段信号转换为多种表征。
|
| 20 |
-
统计和频谱描述整体变化与节律;
|
| 21 |
-
其他分支保留局部形状、相对次序和轴间关系。
|
| 22 |
这里的曲面来自合成信号,是机制示意。
|
| 23 |
|
| 24 |
-
00:01:
|
| 25 |
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 26 |
选定一条截线,就能看到同一窗口内的响应变化。
|
| 27 |
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 28 |
|
| 29 |
-
00:01:
|
| 30 |
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 31 |
再统计各状态出现多少,以及它们如何相互转移。
|
| 32 |
矩阵记录的是转移概率,不是注意力权重。
|
| 33 |
|
| 34 |
-
00:01:
|
| 35 |
几何、区间和形状表征,同时读取轴间结构、局部统计和子序列匹配。
|
| 36 |
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 37 |
保留下来的线索,包括匹配质量和出现位置。
|
| 38 |
|
| 39 |
-
00:
|
| 40 |
互补线索可以在特征层组合。
|
| 41 |
-
CIS 把卷积特征 R、区间特征 I,
|
| 42 |
-
以及频谱、时间和轴间特征 S,拼接成一个向量。
|
| 43 |
随后由同一个分类器读取,输出各活动的分数。
|
| 44 |
|
| 45 |
-
00:02:
|
| 46 |
-
另一种组合发生在分数层:各成员先独立预测,
|
| 47 |
-
再按相同的活动顺序对齐。
|
| 48 |
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 49 |
这与特征拼接,是两种不同的组合方式。
|
| 50 |
|
| 51 |
-
00:02:
|
| 52 |
-
在冻结的论文评估范围内,
|
| 53 |
-
|
| 54 |
-
十八个外部基线中,最高的 RDST 为百分之五十九点六二,
|
| 55 |
-
相差十点五八个百分点。
|
| 56 |
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 57 |
|
| 58 |
-
00:
|
| 59 |
受控消融进一步检验互补性。
|
| 60 |
-
当另外两组特征已经存在时,加入 R 和 S,
|
| 61 |
-
分别平均增加四点八零和三点零一个百分点。
|
| 62 |
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 63 |
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 64 |
|
| 65 |
-
00:
|
| 66 |
-
WristHARBench 提供基准数据;
|
| 67 |
-
WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
| 68 |
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 69 |
-
从公开入口,继续探索。
|
|
|
|
| 1 |
00:00:00,000 从腕部信号,读懂活动
|
| 2 |
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 3 |
+
局部形状、节律、先后次序,还有不同轴之间的关系,都可能帮助识别活动。
|
|
|
|
| 4 |
|
| 5 |
+
00:00:14,400 WristHARBench:统一比较
|
| 6 |
+
如果把身体不同部位的传感器混在一起,结果就难以回答:只看手腕,究竟能识别什么?
|
|
|
|
| 7 |
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 8 |
另有两个多通道惯性数据集,单独分析。
|
| 9 |
|
| 10 |
+
00:00:32,133 先按人划分,再比较模型
|
| 11 |
+
每次划分中,不同参与者进入训练、验证和测试组,同一个人不会跨组出现。
|
|
|
|
| 12 |
模型设置只用验证组选择,测试组留到最后。
|
| 13 |
所有方法使用相同的窗口和划分,重复五次。
|
| 14 |
|
| 15 |
+
00:00:49,967 同一窗口,多种读法
|
| 16 |
WISP 把同一段信号转换为多种表征。
|
| 17 |
+
统计和频谱描述整体变化与节律;其他分支保留局部形状、相对次序和轴间关系。
|
|
|
|
| 18 |
这里的曲面来自合成信号,是机制示意。
|
| 19 |
|
| 20 |
+
00:01:11,233 RC:响应如何变成特征
|
| 21 |
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 22 |
选定一条截线,就能看到同一窗口内的响应变化。
|
| 23 |
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 24 |
|
| 25 |
+
00:01:30,567 SO:保留相对状态与次序
|
| 26 |
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 27 |
再统计各状态出现多少,以及它们如何相互转移。
|
| 28 |
矩阵记录的是转移概率,不是注意力权重。
|
| 29 |
|
| 30 |
+
00:01:49,500 GIS:在窗口里寻找形状
|
| 31 |
几何、区间和形状表征,同时读取轴间结构、局部统计和子序列匹配。
|
| 32 |
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 33 |
保留下来的线索,包括匹配质量和出现位置。
|
| 34 |
|
| 35 |
+
00:02:08,167 CIS:先拼接特征
|
| 36 |
互补线索可以在特征层组合。
|
| 37 |
+
CIS 把卷积特征 R、区间特征 I,以及频谱、时间和轴间特征 S,拼接成一个向量。
|
|
|
|
| 38 |
随后由同一个分类器读取,输出各活动的分数。
|
| 39 |
|
| 40 |
+
00:02:26,267 CSE / ESE:再融合分数
|
| 41 |
+
另一种组合发生在分数层:各成员先独立预测,再按相同的活动顺序对齐。
|
|
|
|
| 42 |
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 43 |
这与特征拼接,是两种不同的组合方式。
|
| 44 |
|
| 45 |
+
00:02:43,767 九个数据集,等权平均
|
| 46 |
+
在冻结的论文评估范围内,随机搜索版 WISP 的平均 Macro F1 为百分之七十点二零。
|
| 47 |
+
十八个外部基线中,最高的 RDST 为百分之五十九点六二,相差十点五八个百分点。
|
|
|
|
|
|
|
| 48 |
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 49 |
|
| 50 |
+
00:03:06,733 增加什么信息,比增加多少更关键
|
| 51 |
受控消融进一步检验互补性。
|
| 52 |
+
当另外两组特征已经存在时,加入 R 和 S,分别平均增加四点八零和三点零一个百分点。
|
|
|
|
| 53 |
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 54 |
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 55 |
|
| 56 |
+
00:03:30,967 从统一评估,到可组合表征
|
| 57 |
+
WristHARBench 提供基准数据;WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
|
|
|
| 58 |
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 59 |
+
从公开入口,继续探索。
|
videos/WISP_zh_system_voice.mp4
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a06a03c57cfc7e2cd6936fd232a7e4ee931255250eb08a4a16eff32635a1f8c1
|
| 3 |
+
size 105702723
|
videos/archive/20261023/WISP_en.srt
ADDED
|
@@ -0,0 +1,235 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
1
|
| 2 |
+
00:00:00,450 --> 00:00:02,424
|
| 3 |
+
Walking, cooking, and handling objects
|
| 4 |
+
|
| 5 |
+
2
|
| 6 |
+
00:00:02,424 --> 00:00:04,398
|
| 7 |
+
leave different clues in wrist motion.
|
| 8 |
+
|
| 9 |
+
3
|
| 10 |
+
00:00:04,638 --> 00:00:07,042
|
| 11 |
+
Local shape, rhythm, temporal order, and
|
| 12 |
+
|
| 13 |
+
4
|
| 14 |
+
00:00:07,042 --> 00:00:10,709
|
| 15 |
+
relationships between axes can all help identify an activity.
|
| 16 |
+
|
| 17 |
+
5
|
| 18 |
+
00:00:13,450 --> 00:00:15,595
|
| 19 |
+
When benchmarks mix sensor locations, they
|
| 20 |
+
|
| 21 |
+
6
|
| 22 |
+
00:00:15,595 --> 00:00:17,843
|
| 23 |
+
cannot isolate what the wrist alone reveals.
|
| 24 |
+
|
| 25 |
+
7
|
| 26 |
+
00:00:18,083 --> 00:00:20,178
|
| 27 |
+
WristHARBench brings together nine wrist
|
| 28 |
+
|
| 29 |
+
8
|
| 30 |
+
00:00:20,178 --> 00:00:22,902
|
| 31 |
+
acceleration datasets under one evaluation contract.
|
| 32 |
+
|
| 33 |
+
9
|
| 34 |
+
00:00:23,142 --> 00:00:25,031
|
| 35 |
+
Two additional multichannel inertial
|
| 36 |
+
|
| 37 |
+
10
|
| 38 |
+
00:00:25,031 --> 00:00:27,602
|
| 39 |
+
datasets support a separate sensitivity analysis.
|
| 40 |
+
|
| 41 |
+
11
|
| 42 |
+
00:00:30,450 --> 00:00:32,538
|
| 43 |
+
Within each repeat, different people
|
| 44 |
+
|
| 45 |
+
12
|
| 46 |
+
00:00:32,538 --> 00:00:35,091
|
| 47 |
+
enter training, validation, and test groups.
|
| 48 |
+
|
| 49 |
+
13
|
| 50 |
+
00:00:35,331 --> 00:00:37,231
|
| 51 |
+
No person appears in two groups.
|
| 52 |
+
|
| 53 |
+
14
|
| 54 |
+
00:00:37,231 --> 00:00:40,259
|
| 55 |
+
Validation guides model choice; testing comes last.
|
| 56 |
+
|
| 57 |
+
15
|
| 58 |
+
00:00:40,499 --> 00:00:44,406
|
| 59 |
+
Methods share the same windows and splits, across five repeated holdouts.
|
| 60 |
+
|
| 61 |
+
16
|
| 62 |
+
00:00:46,450 --> 00:00:50,120
|
| 63 |
+
WISP reads the same signal window through complementary representations.
|
| 64 |
+
|
| 65 |
+
17
|
| 66 |
+
00:00:50,360 --> 00:00:53,740
|
| 67 |
+
Statistics and spectra describe overall variation and rhythm;
|
| 68 |
+
|
| 69 |
+
18
|
| 70 |
+
00:00:53,740 --> 00:00:57,121
|
| 71 |
+
other branches retain shape, order, and cross-axis structure.
|
| 72 |
+
|
| 73 |
+
19
|
| 74 |
+
00:00:57,361 --> 00:01:00,377
|
| 75 |
+
These landscapes illustrate operators on synthetic signals.
|
| 76 |
+
|
| 77 |
+
20
|
| 78 |
+
00:01:01,683 --> 00:01:03,816
|
| 79 |
+
Random convolution kernels move through
|
| 80 |
+
|
| 81 |
+
21
|
| 82 |
+
00:01:03,816 --> 00:01:06,223
|
| 83 |
+
time, responding to local waveform patterns.
|
| 84 |
+
|
| 85 |
+
22
|
| 86 |
+
00:01:06,463 --> 00:01:09,757
|
| 87 |
+
A selected slice reveals the responses within one signal window.
|
| 88 |
+
|
| 89 |
+
23
|
| 90 |
+
00:01:09,997 --> 00:01:12,797
|
| 91 |
+
Positive proportion, maximum, mean, and standard
|
| 92 |
+
|
| 93 |
+
24
|
| 94 |
+
00:01:12,797 --> 00:01:15,597
|
| 95 |
+
deviation pool that response into four features.
|
| 96 |
+
|
| 97 |
+
25
|
| 98 |
+
00:01:19,683 --> 00:01:21,788
|
| 99 |
+
The state-and-order representation maps
|
| 100 |
+
|
| 101 |
+
26
|
| 102 |
+
00:01:21,788 --> 00:01:24,487
|
| 103 |
+
values to relative, within-window quantile states.
|
| 104 |
+
|
| 105 |
+
27
|
| 106 |
+
00:01:24,727 --> 00:01:27,870
|
| 107 |
+
It counts how often states occur and how they transition to each other.
|
| 108 |
+
|
| 109 |
+
28
|
| 110 |
+
00:01:28,110 --> 00:01:31,806
|
| 111 |
+
The matrix contains transition probabilities, not attention weights.
|
| 112 |
+
|
| 113 |
+
29
|
| 114 |
+
00:01:33,683 --> 00:01:36,487
|
| 115 |
+
Geometry, interval, and shape features combine
|
| 116 |
+
|
| 117 |
+
30
|
| 118 |
+
00:01:36,487 --> 00:01:40,389
|
| 119 |
+
cross-axis structure, local statistics, and subsequence matches.
|
| 120 |
+
|
| 121 |
+
31
|
| 122 |
+
00:01:40,629 --> 00:01:42,633
|
| 123 |
+
A template moves along the window. The
|
| 124 |
+
|
| 125 |
+
32
|
| 126 |
+
00:01:42,633 --> 00:01:44,847
|
| 127 |
+
lowest distance identifies its best match.
|
| 128 |
+
|
| 129 |
+
33
|
| 130 |
+
00:01:45,087 --> 00:01:47,587
|
| 131 |
+
Both match quality and location become features.
|
| 132 |
+
|
| 133 |
+
34
|
| 134 |
+
00:01:48,883 --> 00:01:51,624
|
| 135 |
+
Complementary clues can be combined at the feature level.
|
| 136 |
+
|
| 137 |
+
35
|
| 138 |
+
00:01:51,864 --> 00:01:55,855
|
| 139 |
+
features I, and spectral, temporal, and cross-axis features S.
|
| 140 |
+
|
| 141 |
+
36
|
| 142 |
+
00:01:55,855 --> 00:01:59,073
|
| 143 |
+
features I, and spectral, temporal, and cross-axis
|
| 144 |
+
|
| 145 |
+
37
|
| 146 |
+
00:01:59,313 --> 00:02:03,210
|
| 147 |
+
A single classifier reads the combined vector and produces activity scores.
|
| 148 |
+
|
| 149 |
+
38
|
| 150 |
+
00:02:04,883 --> 00:02:07,104
|
| 151 |
+
Score ensembles combine independently fitted
|
| 152 |
+
|
| 153 |
+
39
|
| 154 |
+
00:02:07,104 --> 00:02:09,425
|
| 155 |
+
members after aligning their activity classes.
|
| 156 |
+
|
| 157 |
+
40
|
| 158 |
+
00:02:09,665 --> 00:02:13,544
|
| 159 |
+
CSE averages three members equally; ESE uses weighted averaging.
|
| 160 |
+
|
| 161 |
+
41
|
| 162 |
+
00:02:13,784 --> 00:02:17,308
|
| 163 |
+
This is different from concatenating features before one classifier.
|
| 164 |
+
|
| 165 |
+
42
|
| 166 |
+
00:02:19,883 --> 00:02:22,641
|
| 167 |
+
On the frozen paper evaluation, WISP with
|
| 168 |
+
|
| 169 |
+
43
|
| 170 |
+
00:02:22,641 --> 00:02:26,004
|
| 171 |
+
random search reaches 70.20 percent mean macro F1.
|
| 172 |
+
|
| 173 |
+
44
|
| 174 |
+
00:02:26,244 --> 00:02:29,088
|
| 175 |
+
RDST, the strongest of eighteen external
|
| 176 |
+
|
| 177 |
+
45
|
| 178 |
+
00:02:29,088 --> 00:02:32,571
|
| 179 |
+
baselines on this average, reaches 59.62 percent.
|
| 180 |
+
|
| 181 |
+
46
|
| 182 |
+
00:02:32,811 --> 00:02:35,657
|
| 183 |
+
The difference is 10.58 percentage points. This
|
| 184 |
+
|
| 185 |
+
47
|
| 186 |
+
00:02:35,657 --> 00:02:38,623
|
| 187 |
+
is a dataset average, not a win on every dataset.
|
| 188 |
+
|
| 189 |
+
48
|
| 190 |
+
00:02:39,917 --> 00:02:43,940
|
| 191 |
+
Controlled ablations test complementary information with a fixed classifier.
|
| 192 |
+
|
| 193 |
+
49
|
| 194 |
+
00:02:44,180 --> 00:02:47,181
|
| 195 |
+
With the other two groups already present,
|
| 196 |
+
|
| 197 |
+
50
|
| 198 |
+
00:02:47,181 --> 00:02:50,610
|
| 199 |
+
adding R gains 4.80 points; adding S gains 3.01.
|
| 200 |
+
|
| 201 |
+
51
|
| 202 |
+
00:02:50,850 --> 00:02:53,367
|
| 203 |
+
Adding I changes the mean by minus 0.27
|
| 204 |
+
|
| 205 |
+
52
|
| 206 |
+
00:02:53,367 --> 00:02:55,883
|
| 207 |
+
points, with an interval crossing zero.
|
| 208 |
+
|
| 209 |
+
53
|
| 210 |
+
00:02:56,123 --> 00:02:58,322
|
| 211 |
+
The evidence supports adding missing clues,
|
| 212 |
+
|
| 213 |
+
54
|
| 214 |
+
00:02:58,322 --> 00:03:00,623
|
| 215 |
+
rather than more overlapping representations.
|
| 216 |
+
|
| 217 |
+
55
|
| 218 |
+
00:03:01,917 --> 00:03:04,910
|
| 219 |
+
WristHARBench provides benchmark data. WISP provides
|
| 220 |
+
|
| 221 |
+
56
|
| 222 |
+
00:03:04,910 --> 00:03:08,076
|
| 223 |
+
method code, saved results, and retained search models.
|
| 224 |
+
|
| 225 |
+
57
|
| 226 |
+
00:03:08,316 --> 00:03:10,712
|
| 227 |
+
The 360 retained checkpoints do not cover
|
| 228 |
+
|
| 229 |
+
58
|
| 230 |
+
00:03:10,712 --> 00:03:13,516
|
| 231 |
+
every method or imply complete inference reruns.
|
| 232 |
+
|
| 233 |
+
59
|
| 234 |
+
00:03:13,756 --> 00:03:15,325
|
| 235 |
+
Explore the public releases.
|
videos/archive/20261023/WISP_en.vtt
ADDED
|
@@ -0,0 +1,178 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
WEBVTT
|
| 2 |
+
|
| 3 |
+
00:00:00.450 --> 00:00:02.424
|
| 4 |
+
Walking, cooking, and handling objects
|
| 5 |
+
|
| 6 |
+
00:00:02.424 --> 00:00:04.398
|
| 7 |
+
leave different clues in wrist motion.
|
| 8 |
+
|
| 9 |
+
00:00:04.638 --> 00:00:07.042
|
| 10 |
+
Local shape, rhythm, temporal order, and
|
| 11 |
+
|
| 12 |
+
00:00:07.042 --> 00:00:10.709
|
| 13 |
+
relationships between axes can all help identify an activity.
|
| 14 |
+
|
| 15 |
+
00:00:13.450 --> 00:00:15.595
|
| 16 |
+
When benchmarks mix sensor locations, they
|
| 17 |
+
|
| 18 |
+
00:00:15.595 --> 00:00:17.843
|
| 19 |
+
cannot isolate what the wrist alone reveals.
|
| 20 |
+
|
| 21 |
+
00:00:18.083 --> 00:00:20.178
|
| 22 |
+
WristHARBench brings together nine wrist
|
| 23 |
+
|
| 24 |
+
00:00:20.178 --> 00:00:22.902
|
| 25 |
+
acceleration datasets under one evaluation contract.
|
| 26 |
+
|
| 27 |
+
00:00:23.142 --> 00:00:25.031
|
| 28 |
+
Two additional multichannel inertial
|
| 29 |
+
|
| 30 |
+
00:00:25.031 --> 00:00:27.602
|
| 31 |
+
datasets support a separate sensitivity analysis.
|
| 32 |
+
|
| 33 |
+
00:00:30.450 --> 00:00:32.538
|
| 34 |
+
Within each repeat, different people
|
| 35 |
+
|
| 36 |
+
00:00:32.538 --> 00:00:35.091
|
| 37 |
+
enter training, validation, and test groups.
|
| 38 |
+
|
| 39 |
+
00:00:35.331 --> 00:00:37.231
|
| 40 |
+
No person appears in two groups.
|
| 41 |
+
|
| 42 |
+
00:00:37.231 --> 00:00:40.259
|
| 43 |
+
Validation guides model choice; testing comes last.
|
| 44 |
+
|
| 45 |
+
00:00:40.499 --> 00:00:44.406
|
| 46 |
+
Methods share the same windows and splits, across five repeated holdouts.
|
| 47 |
+
|
| 48 |
+
00:00:46.450 --> 00:00:50.120
|
| 49 |
+
WISP reads the same signal window through complementary representations.
|
| 50 |
+
|
| 51 |
+
00:00:50.360 --> 00:00:53.740
|
| 52 |
+
Statistics and spectra describe overall variation and rhythm;
|
| 53 |
+
|
| 54 |
+
00:00:53.740 --> 00:00:57.121
|
| 55 |
+
other branches retain shape, order, and cross-axis structure.
|
| 56 |
+
|
| 57 |
+
00:00:57.361 --> 00:01:00.377
|
| 58 |
+
These landscapes illustrate operators on synthetic signals.
|
| 59 |
+
|
| 60 |
+
00:01:01.683 --> 00:01:03.816
|
| 61 |
+
Random convolution kernels move through
|
| 62 |
+
|
| 63 |
+
00:01:03.816 --> 00:01:06.223
|
| 64 |
+
time, responding to local waveform patterns.
|
| 65 |
+
|
| 66 |
+
00:01:06.463 --> 00:01:09.757
|
| 67 |
+
A selected slice reveals the responses within one signal window.
|
| 68 |
+
|
| 69 |
+
00:01:09.997 --> 00:01:12.797
|
| 70 |
+
Positive proportion, maximum, mean, and standard
|
| 71 |
+
|
| 72 |
+
00:01:12.797 --> 00:01:15.597
|
| 73 |
+
deviation pool that response into four features.
|
| 74 |
+
|
| 75 |
+
00:01:19.683 --> 00:01:21.788
|
| 76 |
+
The state-and-order representation maps
|
| 77 |
+
|
| 78 |
+
00:01:21.788 --> 00:01:24.487
|
| 79 |
+
values to relative, within-window quantile states.
|
| 80 |
+
|
| 81 |
+
00:01:24.727 --> 00:01:27.870
|
| 82 |
+
It counts how often states occur and how they transition to each other.
|
| 83 |
+
|
| 84 |
+
00:01:28.110 --> 00:01:31.806
|
| 85 |
+
The matrix contains transition probabilities, not attention weights.
|
| 86 |
+
|
| 87 |
+
00:01:33.683 --> 00:01:36.487
|
| 88 |
+
Geometry, interval, and shape features combine
|
| 89 |
+
|
| 90 |
+
00:01:36.487 --> 00:01:40.389
|
| 91 |
+
cross-axis structure, local statistics, and subsequence matches.
|
| 92 |
+
|
| 93 |
+
00:01:40.629 --> 00:01:42.633
|
| 94 |
+
A template moves along the window. The
|
| 95 |
+
|
| 96 |
+
00:01:42.633 --> 00:01:44.847
|
| 97 |
+
lowest distance identifies its best match.
|
| 98 |
+
|
| 99 |
+
00:01:45.087 --> 00:01:47.587
|
| 100 |
+
Both match quality and location become features.
|
| 101 |
+
|
| 102 |
+
00:01:48.883 --> 00:01:51.624
|
| 103 |
+
Complementary clues can be combined at the feature level.
|
| 104 |
+
|
| 105 |
+
00:01:51.864 --> 00:01:55.855
|
| 106 |
+
features I, and spectral, temporal, and cross-axis features S.
|
| 107 |
+
|
| 108 |
+
00:01:55.855 --> 00:01:59.073
|
| 109 |
+
features I, and spectral, temporal, and cross-axis
|
| 110 |
+
|
| 111 |
+
00:01:59.313 --> 00:02:03.210
|
| 112 |
+
A single classifier reads the combined vector and produces activity scores.
|
| 113 |
+
|
| 114 |
+
00:02:04.883 --> 00:02:07.104
|
| 115 |
+
Score ensembles combine independently fitted
|
| 116 |
+
|
| 117 |
+
00:02:07.104 --> 00:02:09.425
|
| 118 |
+
members after aligning their activity classes.
|
| 119 |
+
|
| 120 |
+
00:02:09.665 --> 00:02:13.544
|
| 121 |
+
CSE averages three members equally; ESE uses weighted averaging.
|
| 122 |
+
|
| 123 |
+
00:02:13.784 --> 00:02:17.308
|
| 124 |
+
This is different from concatenating features before one classifier.
|
| 125 |
+
|
| 126 |
+
00:02:19.883 --> 00:02:22.641
|
| 127 |
+
On the frozen paper evaluation, WISP with
|
| 128 |
+
|
| 129 |
+
00:02:22.641 --> 00:02:26.004
|
| 130 |
+
random search reaches 70.20 percent mean macro F1.
|
| 131 |
+
|
| 132 |
+
00:02:26.244 --> 00:02:29.088
|
| 133 |
+
RDST, the strongest of eighteen external
|
| 134 |
+
|
| 135 |
+
00:02:29.088 --> 00:02:32.571
|
| 136 |
+
baselines on this average, reaches 59.62 percent.
|
| 137 |
+
|
| 138 |
+
00:02:32.811 --> 00:02:35.657
|
| 139 |
+
The difference is 10.58 percentage points. This
|
| 140 |
+
|
| 141 |
+
00:02:35.657 --> 00:02:38.623
|
| 142 |
+
is a dataset average, not a win on every dataset.
|
| 143 |
+
|
| 144 |
+
00:02:39.917 --> 00:02:43.940
|
| 145 |
+
Controlled ablations test complementary information with a fixed classifier.
|
| 146 |
+
|
| 147 |
+
00:02:44.180 --> 00:02:47.181
|
| 148 |
+
With the other two groups already present,
|
| 149 |
+
|
| 150 |
+
00:02:47.181 --> 00:02:50.610
|
| 151 |
+
adding R gains 4.80 points; adding S gains 3.01.
|
| 152 |
+
|
| 153 |
+
00:02:50.850 --> 00:02:53.367
|
| 154 |
+
Adding I changes the mean by minus 0.27
|
| 155 |
+
|
| 156 |
+
00:02:53.367 --> 00:02:55.883
|
| 157 |
+
points, with an interval crossing zero.
|
| 158 |
+
|
| 159 |
+
00:02:56.123 --> 00:02:58.322
|
| 160 |
+
The evidence supports adding missing clues,
|
| 161 |
+
|
| 162 |
+
00:02:58.322 --> 00:03:00.623
|
| 163 |
+
rather than more overlapping representations.
|
| 164 |
+
|
| 165 |
+
00:03:01.917 --> 00:03:04.910
|
| 166 |
+
WristHARBench provides benchmark data. WISP provides
|
| 167 |
+
|
| 168 |
+
00:03:04.910 --> 00:03:08.076
|
| 169 |
+
method code, saved results, and retained search models.
|
| 170 |
+
|
| 171 |
+
00:03:08.316 --> 00:03:10.712
|
| 172 |
+
The 360 retained checkpoints do not cover
|
| 173 |
+
|
| 174 |
+
00:03:10.712 --> 00:03:13.516
|
| 175 |
+
every method or imply complete inference reruns.
|
| 176 |
+
|
| 177 |
+
00:03:13.756 --> 00:03:15.325
|
| 178 |
+
Explore the public releases.
|
videos/archive/20261023/WISP_en_cover.png
ADDED
|
Git LFS Details
|
videos/archive/20261023/WISP_en_script.txt
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
00:00:00,000 Reading activity at the wrist
|
| 2 |
+
Walking, cooking, and handling objects
|
| 3 |
+
leave different clues in wrist motion.
|
| 4 |
+
Local shape, rhythm, temporal order, and
|
| 5 |
+
relationships between axes can all help identify an activity.
|
| 6 |
+
|
| 7 |
+
00:00:13,000 WristHARBench: a common comparison
|
| 8 |
+
When benchmarks mix sensor locations, they
|
| 9 |
+
cannot isolate what the wrist alone reveals.
|
| 10 |
+
WristHARBench brings together nine wrist
|
| 11 |
+
acceleration datasets under one evaluation contract.
|
| 12 |
+
Two additional multichannel inertial
|
| 13 |
+
datasets support a separate sensitivity analysis.
|
| 14 |
+
|
| 15 |
+
00:00:30,000 Separate people before comparing models
|
| 16 |
+
Within each repeat, different people
|
| 17 |
+
enter training, validation, and test groups.
|
| 18 |
+
No person appears in two groups.
|
| 19 |
+
Validation guides model choice; testing comes last.
|
| 20 |
+
Methods share the same windows and splits, across five repeated holdouts.
|
| 21 |
+
|
| 22 |
+
00:00:46,000 One window, several readings
|
| 23 |
+
WISP reads the same signal window through complementary representations.
|
| 24 |
+
Statistics and spectra describe overall variation and rhythm;
|
| 25 |
+
other branches retain shape, order, and cross-axis structure.
|
| 26 |
+
These landscapes illustrate operators on synthetic signals.
|
| 27 |
+
|
| 28 |
+
00:01:01,233 RC: from responses to features
|
| 29 |
+
Random convolution kernels move through
|
| 30 |
+
time, responding to local waveform patterns.
|
| 31 |
+
A selected slice reveals the responses within one signal window.
|
| 32 |
+
Positive proportion, maximum, mean, and standard
|
| 33 |
+
deviation pool that response into four features.
|
| 34 |
+
|
| 35 |
+
00:01:19,233 SO: relative states and their order
|
| 36 |
+
The state-and-order representation maps
|
| 37 |
+
values to relative, within-window quantile states.
|
| 38 |
+
It counts how often states occur and how they transition to each other.
|
| 39 |
+
The matrix contains transition probabilities, not attention weights.
|
| 40 |
+
|
| 41 |
+
00:01:33,233 GIS: finding shapes within a window
|
| 42 |
+
Geometry, interval, and shape features combine
|
| 43 |
+
cross-axis structure, local statistics, and subsequence matches.
|
| 44 |
+
A template moves along the window. The
|
| 45 |
+
lowest distance identifies its best match.
|
| 46 |
+
Both match quality and location become features.
|
| 47 |
+
|
| 48 |
+
00:01:48,433 CIS: concatenate features first
|
| 49 |
+
Complementary clues can be combined at the feature level.
|
| 50 |
+
features I, and spectral, temporal, and cross-axis features S.
|
| 51 |
+
features I, and spectral, temporal, and cross-axis
|
| 52 |
+
A single classifier reads the combined vector and produces activity scores.
|
| 53 |
+
|
| 54 |
+
00:02:04,433 CSE / ESE: combine class scores
|
| 55 |
+
Score ensembles combine independently fitted
|
| 56 |
+
members after aligning their activity classes.
|
| 57 |
+
CSE averages three members equally; ESE uses weighted averaging.
|
| 58 |
+
This is different from concatenating features before one classifier.
|
| 59 |
+
|
| 60 |
+
00:02:19,433 Nine datasets, equal weight
|
| 61 |
+
On the frozen paper evaluation, WISP with
|
| 62 |
+
random search reaches 70.20 percent mean macro F1.
|
| 63 |
+
RDST, the strongest of eighteen external
|
| 64 |
+
baselines on this average, reaches 59.62 percent.
|
| 65 |
+
The difference is 10.58 percentage points. This
|
| 66 |
+
is a dataset average, not a win on every dataset.
|
| 67 |
+
|
| 68 |
+
00:02:39,467 What information does a branch add?
|
| 69 |
+
Controlled ablations test complementary information with a fixed classifier.
|
| 70 |
+
With the other two groups already present,
|
| 71 |
+
adding R gains 4.80 points; adding S gains 3.01.
|
| 72 |
+
Adding I changes the mean by minus 0.27
|
| 73 |
+
points, with an interval crossing zero.
|
| 74 |
+
The evidence supports adding missing clues,
|
| 75 |
+
rather than more overlapping representations.
|
| 76 |
+
|
| 77 |
+
00:03:01,467 From shared evaluation to composable representations
|
| 78 |
+
WristHARBench provides benchmark data. WISP provides
|
| 79 |
+
method code, saved results, and retained search models.
|
| 80 |
+
The 360 retained checkpoints do not cover
|
| 81 |
+
every method or imply complete inference reruns.
|
| 82 |
+
Explore the public releases.
|
videos/archive/20261023/WISP_en_system_voice.mp4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1d36676464b1eecd714802f6be25c770074b682f510a2142db2b55b7e4822c8
|
| 3 |
+
size 90249749
|
videos/archive/20261023/WISP_zh.srt
ADDED
|
@@ -0,0 +1,183 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
1
|
| 2 |
+
00:00:00,450 --> 00:00:05,905
|
| 3 |
+
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 4 |
+
|
| 5 |
+
2
|
| 6 |
+
00:00:06,145 --> 00:00:10,439
|
| 7 |
+
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 8 |
+
|
| 9 |
+
3
|
| 10 |
+
00:00:10,439 --> 00:00:12,229
|
| 11 |
+
都可能帮助识别活动。
|
| 12 |
+
|
| 13 |
+
4
|
| 14 |
+
00:00:13,550 --> 00:00:19,077
|
| 15 |
+
如果把身体不同部位的传感器混在一起,结果就难以回答:只看手腕,
|
| 16 |
+
|
| 17 |
+
5
|
| 18 |
+
00:00:19,077 --> 00:00:20,503
|
| 19 |
+
究竟能识别什么?
|
| 20 |
+
|
| 21 |
+
6
|
| 22 |
+
00:00:20,743 --> 00:00:24,161
|
| 23 |
+
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 24 |
+
|
| 25 |
+
7
|
| 26 |
+
00:00:24,401 --> 00:00:27,664
|
| 27 |
+
另有两个多通道惯性数据集,单独分析。
|
| 28 |
+
|
| 29 |
+
8
|
| 30 |
+
00:00:30,550 --> 00:00:34,744
|
| 31 |
+
每次划分中,不同参与者进入训练、验证和测试组,
|
| 32 |
+
|
| 33 |
+
9
|
| 34 |
+
00:00:34,744 --> 00:00:36,750
|
| 35 |
+
同一个人不会跨组出现。
|
| 36 |
+
|
| 37 |
+
10
|
| 38 |
+
00:00:36,990 --> 00:00:40,532
|
| 39 |
+
模型设置只用验证组选择,测试组留到最后。
|
| 40 |
+
|
| 41 |
+
11
|
| 42 |
+
00:00:40,772 --> 00:00:44,503
|
| 43 |
+
所有方法使用相同的窗口和划分,重复五次。
|
| 44 |
+
|
| 45 |
+
12
|
| 46 |
+
00:00:46,550 --> 00:00:49,420
|
| 47 |
+
WISP 把同一段信号转换为多种表征。
|
| 48 |
+
|
| 49 |
+
13
|
| 50 |
+
00:00:49,660 --> 00:00:52,435
|
| 51 |
+
统计和频谱描述整体变化与节律;
|
| 52 |
+
|
| 53 |
+
14
|
| 54 |
+
00:00:52,435 --> 00:00:56,319
|
| 55 |
+
其他分支保留局部形状、相对次序和轴间关系。
|
| 56 |
+
|
| 57 |
+
15
|
| 58 |
+
00:00:56,559 --> 00:00:59,751
|
| 59 |
+
这里的曲面来自合成信号,是机制示意。
|
| 60 |
+
|
| 61 |
+
16
|
| 62 |
+
00:01:01,050 --> 00:01:05,351
|
| 63 |
+
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 64 |
+
|
| 65 |
+
17
|
| 66 |
+
00:01:05,591 --> 00:01:09,659
|
| 67 |
+
选定一条截线,就能看到同一窗口内的响应变化。
|
| 68 |
+
|
| 69 |
+
18
|
| 70 |
+
00:01:09,899 --> 00:01:15,345
|
| 71 |
+
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 72 |
+
|
| 73 |
+
19
|
| 74 |
+
00:01:19,050 --> 00:01:23,868
|
| 75 |
+
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 76 |
+
|
| 77 |
+
20
|
| 78 |
+
00:01:24,108 --> 00:01:28,133
|
| 79 |
+
再统计各状态出现多少,以及它们如何相互转移。
|
| 80 |
+
|
| 81 |
+
21
|
| 82 |
+
00:01:28,373 --> 00:01:31,379
|
| 83 |
+
矩阵记录的是转移概率,不是注意力权重。
|
| 84 |
+
|
| 85 |
+
22
|
| 86 |
+
00:01:33,050 --> 00:01:38,900
|
| 87 |
+
几何、区间和形状表征,同时读取轴间结构、局部统计和子序列匹配。
|
| 88 |
+
|
| 89 |
+
23
|
| 90 |
+
00:01:39,140 --> 00:01:42,769
|
| 91 |
+
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 92 |
+
|
| 93 |
+
24
|
| 94 |
+
00:01:43,009 --> 00:01:46,571
|
| 95 |
+
保留下来的线索,包括匹配质量和出现位置。
|
| 96 |
+
|
| 97 |
+
25
|
| 98 |
+
00:01:47,883 --> 00:01:50,102
|
| 99 |
+
互补线索可以在特征层组合。
|
| 100 |
+
|
| 101 |
+
26
|
| 102 |
+
00:01:50,342 --> 00:01:53,625
|
| 103 |
+
CIS 把卷积特征 R、区间特征 I,
|
| 104 |
+
|
| 105 |
+
27
|
| 106 |
+
00:01:53,625 --> 00:01:57,600
|
| 107 |
+
以及频谱、时间和轴间特征 S,拼接成一个向量。
|
| 108 |
+
|
| 109 |
+
28
|
| 110 |
+
00:01:57,840 --> 00:02:01,451
|
| 111 |
+
随后由同一个分类器读取,输出各活动的分数。
|
| 112 |
+
|
| 113 |
+
29
|
| 114 |
+
00:02:03,883 --> 00:02:07,747
|
| 115 |
+
另一种组合发生在分数层:各成员先独立预测,
|
| 116 |
+
|
| 117 |
+
30
|
| 118 |
+
00:02:07,747 --> 00:02:09,955
|
| 119 |
+
再按相同的活动顺序对齐。
|
| 120 |
+
|
| 121 |
+
31
|
| 122 |
+
00:02:10,195 --> 00:02:14,525
|
| 123 |
+
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 124 |
+
|
| 125 |
+
32
|
| 126 |
+
00:02:14,765 --> 00:02:17,912
|
| 127 |
+
这与特征拼接,是两种不同的组合方式。
|
| 128 |
+
|
| 129 |
+
33
|
| 130 |
+
00:02:19,217 --> 00:02:20,903
|
| 131 |
+
在冻结的论文评估范围内,
|
| 132 |
+
|
| 133 |
+
34
|
| 134 |
+
00:02:20,903 --> 00:02:25,682
|
| 135 |
+
随机搜索版 WISP 的平均 Macro F1 为百分之七十点二零。
|
| 136 |
+
|
| 137 |
+
35
|
| 138 |
+
00:02:25,922 --> 00:02:30,753
|
| 139 |
+
十八个外部基线中,最高的 RDST 为百分之五十九点六二,
|
| 140 |
+
|
| 141 |
+
36
|
| 142 |
+
00:02:30,753 --> 00:02:32,586
|
| 143 |
+
相差十点五八个百分点。
|
| 144 |
+
|
| 145 |
+
37
|
| 146 |
+
00:02:32,826 --> 00:02:37,090
|
| 147 |
+
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 148 |
+
|
| 149 |
+
38
|
| 150 |
+
00:02:38,383 --> 00:02:40,657
|
| 151 |
+
受控消融进一步检验互补性。
|
| 152 |
+
|
| 153 |
+
39
|
| 154 |
+
00:02:40,897 --> 00:02:44,778
|
| 155 |
+
当另外两组特征已经存在时,加入 R 和 S,
|
| 156 |
+
|
| 157 |
+
40
|
| 158 |
+
00:02:44,778 --> 00:02:48,307
|
| 159 |
+
分别平均增加四点八零和三点零一个百分点。
|
| 160 |
+
|
| 161 |
+
41
|
| 162 |
+
00:02:48,547 --> 00:02:52,390
|
| 163 |
+
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 164 |
+
|
| 165 |
+
42
|
| 166 |
+
00:02:52,630 --> 00:02:56,893
|
| 167 |
+
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 168 |
+
|
| 169 |
+
43
|
| 170 |
+
00:02:58,217 --> 00:03:01,034
|
| 171 |
+
WristHARBench 提供基准数据;
|
| 172 |
+
|
| 173 |
+
44
|
| 174 |
+
00:03:01,034 --> 00:03:04,656
|
| 175 |
+
WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
| 176 |
+
|
| 177 |
+
45
|
| 178 |
+
00:03:04,896 --> 00:03:09,297
|
| 179 |
+
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 180 |
+
|
| 181 |
+
46
|
| 182 |
+
00:03:09,537 --> 00:03:11,598
|
| 183 |
+
从公开入口,继续探索。
|
videos/archive/20261023/WISP_zh.vtt
ADDED
|
@@ -0,0 +1,139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
WEBVTT
|
| 2 |
+
|
| 3 |
+
00:00:00.450 --> 00:00:05.905
|
| 4 |
+
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 5 |
+
|
| 6 |
+
00:00:06.145 --> 00:00:10.439
|
| 7 |
+
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 8 |
+
|
| 9 |
+
00:00:10.439 --> 00:00:12.229
|
| 10 |
+
都可能帮助识别活动。
|
| 11 |
+
|
| 12 |
+
00:00:13.550 --> 00:00:19.077
|
| 13 |
+
如果把身体不同部位的传感器混在一起,结果就难以回答:只看手腕,
|
| 14 |
+
|
| 15 |
+
00:00:19.077 --> 00:00:20.503
|
| 16 |
+
究竟能识别什么?
|
| 17 |
+
|
| 18 |
+
00:00:20.743 --> 00:00:24.161
|
| 19 |
+
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 20 |
+
|
| 21 |
+
00:00:24.401 --> 00:00:27.664
|
| 22 |
+
另有两个多通道惯性数据集,单独分析。
|
| 23 |
+
|
| 24 |
+
00:00:30.550 --> 00:00:34.744
|
| 25 |
+
每次划分中,不同参与者进入训练、验证和测试组,
|
| 26 |
+
|
| 27 |
+
00:00:34.744 --> 00:00:36.750
|
| 28 |
+
同一个人不会跨组出现。
|
| 29 |
+
|
| 30 |
+
00:00:36.990 --> 00:00:40.532
|
| 31 |
+
模型设置只用验证组选择,测试组留到最后。
|
| 32 |
+
|
| 33 |
+
00:00:40.772 --> 00:00:44.503
|
| 34 |
+
所有方法使用相同的窗口和划分,重复五次。
|
| 35 |
+
|
| 36 |
+
00:00:46.550 --> 00:00:49.420
|
| 37 |
+
WISP 把同一段信号转换为多种表征。
|
| 38 |
+
|
| 39 |
+
00:00:49.660 --> 00:00:52.435
|
| 40 |
+
统计和频谱描述整体变化与节律;
|
| 41 |
+
|
| 42 |
+
00:00:52.435 --> 00:00:56.319
|
| 43 |
+
其他分支保留局部形状、相对次序和轴间关系。
|
| 44 |
+
|
| 45 |
+
00:00:56.559 --> 00:00:59.751
|
| 46 |
+
这里的曲面来自合成信号,是机制示意。
|
| 47 |
+
|
| 48 |
+
00:01:01.050 --> 00:01:05.351
|
| 49 |
+
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 50 |
+
|
| 51 |
+
00:01:05.591 --> 00:01:09.659
|
| 52 |
+
选定一条截线,就能看到同一窗口内的响应变化。
|
| 53 |
+
|
| 54 |
+
00:01:09.899 --> 00:01:15.345
|
| 55 |
+
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 56 |
+
|
| 57 |
+
00:01:19.050 --> 00:01:23.868
|
| 58 |
+
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 59 |
+
|
| 60 |
+
00:01:24.108 --> 00:01:28.133
|
| 61 |
+
再统计各状态出现多少,以及它们如何相互转移。
|
| 62 |
+
|
| 63 |
+
00:01:28.373 --> 00:01:31.379
|
| 64 |
+
矩阵记录的是转移概率,不是注意力权重。
|
| 65 |
+
|
| 66 |
+
00:01:33.050 --> 00:01:38.900
|
| 67 |
+
几何、区间和形状表征,同时读取轴间结构、局部统计和子序列匹配。
|
| 68 |
+
|
| 69 |
+
00:01:39.140 --> 00:01:42.769
|
| 70 |
+
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 71 |
+
|
| 72 |
+
00:01:43.009 --> 00:01:46.571
|
| 73 |
+
保留下来的线索,包括匹配质量和出现位置。
|
| 74 |
+
|
| 75 |
+
00:01:47.883 --> 00:01:50.102
|
| 76 |
+
互补线索可以在特征层组合。
|
| 77 |
+
|
| 78 |
+
00:01:50.342 --> 00:01:53.625
|
| 79 |
+
CIS 把卷积特征 R、区间特征 I,
|
| 80 |
+
|
| 81 |
+
00:01:53.625 --> 00:01:57.600
|
| 82 |
+
以及频谱、时间和轴间特征 S,拼接成一个向量。
|
| 83 |
+
|
| 84 |
+
00:01:57.840 --> 00:02:01.451
|
| 85 |
+
随后由同一个分类器读取,输出各活动的分数。
|
| 86 |
+
|
| 87 |
+
00:02:03.883 --> 00:02:07.747
|
| 88 |
+
另一种组合发生在分数层:各成员先独立预测,
|
| 89 |
+
|
| 90 |
+
00:02:07.747 --> 00:02:09.955
|
| 91 |
+
再按相同的活动顺序对齐。
|
| 92 |
+
|
| 93 |
+
00:02:10.195 --> 00:02:14.525
|
| 94 |
+
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 95 |
+
|
| 96 |
+
00:02:14.765 --> 00:02:17.912
|
| 97 |
+
这与特征拼接,是两种不同的组合方式。
|
| 98 |
+
|
| 99 |
+
00:02:19.217 --> 00:02:20.903
|
| 100 |
+
在冻结的论文评估范围内,
|
| 101 |
+
|
| 102 |
+
00:02:20.903 --> 00:02:25.682
|
| 103 |
+
随机搜索版 WISP 的平均 Macro F1 为百分之七十点二零。
|
| 104 |
+
|
| 105 |
+
00:02:25.922 --> 00:02:30.753
|
| 106 |
+
十八个外部基线中,最高的 RDST 为百分之五十九点六二,
|
| 107 |
+
|
| 108 |
+
00:02:30.753 --> 00:02:32.586
|
| 109 |
+
相差十点五八个百分点。
|
| 110 |
+
|
| 111 |
+
00:02:32.826 --> 00:02:37.090
|
| 112 |
+
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 113 |
+
|
| 114 |
+
00:02:38.383 --> 00:02:40.657
|
| 115 |
+
受控消融进一步检验互补性。
|
| 116 |
+
|
| 117 |
+
00:02:40.897 --> 00:02:44.778
|
| 118 |
+
当另外两组特征已经存在时,加入 R 和 S,
|
| 119 |
+
|
| 120 |
+
00:02:44.778 --> 00:02:48.307
|
| 121 |
+
分别平均增加四点八零和三点零一个百分点。
|
| 122 |
+
|
| 123 |
+
00:02:48.547 --> 00:02:52.390
|
| 124 |
+
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 125 |
+
|
| 126 |
+
00:02:52.630 --> 00:02:56.893
|
| 127 |
+
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 128 |
+
|
| 129 |
+
00:02:58.217 --> 00:03:01.034
|
| 130 |
+
WristHARBench 提供基准数据;
|
| 131 |
+
|
| 132 |
+
00:03:01.034 --> 00:03:04.656
|
| 133 |
+
WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
| 134 |
+
|
| 135 |
+
00:03:04.896 --> 00:03:09.297
|
| 136 |
+
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 137 |
+
|
| 138 |
+
00:03:09.537 --> 00:03:11.598
|
| 139 |
+
从公开入口,继续探索。
|
videos/archive/20261023/WISP_zh_cover.png
ADDED
|
Git LFS Details
|
videos/archive/20261023/WISP_zh_script.txt
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
00:00:00,000 从腕部信号,读懂活动
|
| 2 |
+
同样是手腕运动,走路、做饭和整理物品,留下的线索并不相同。
|
| 3 |
+
局部形状、节律、先后次序,还有不同轴之间的关系,
|
| 4 |
+
都可能帮助识别活动。
|
| 5 |
+
|
| 6 |
+
00:00:13,100 WristHARBench:统一比较
|
| 7 |
+
如果把身体不同部位的传感器混在一起,结果就难以回答:只看手腕,
|
| 8 |
+
究竟能识别什么?
|
| 9 |
+
WristHARBench 整理了九个腕部三轴加速度数据集。
|
| 10 |
+
另有两个多通道惯性数据集,单独分析。
|
| 11 |
+
|
| 12 |
+
00:00:30,100 先按人划分,再比较模型
|
| 13 |
+
每次划分中,不同参与者进入训练、验证和测试组,
|
| 14 |
+
同一个人不会跨组出现。
|
| 15 |
+
模型设置只用验证组选择,测试组留到最后。
|
| 16 |
+
所有方法使用相同的窗口和划分,重复五次。
|
| 17 |
+
|
| 18 |
+
00:00:46,100 同一窗口,多种读法
|
| 19 |
+
WISP 把同一段信号转换为多种表征。
|
| 20 |
+
统计和频谱描述整体变化与节律;
|
| 21 |
+
其他分支保留局部形状、相对次序和轴间关系。
|
| 22 |
+
这里的曲面来自合成信号,是机制示意。
|
| 23 |
+
|
| 24 |
+
00:01:00,600 RC:响应如何变成特征
|
| 25 |
+
随机卷积核沿时间移动,对不同的局部波形产生响应。
|
| 26 |
+
选定一条截线,就能看到同一窗口内的响应变化。
|
| 27 |
+
正值比例、最大值、均值和标准差,把整条响应汇总为四个特征。
|
| 28 |
+
|
| 29 |
+
00:01:18,600 SO:保留相对状态与次序
|
| 30 |
+
状态与次序表征,把窗口内的数值按分位数转成相对状态。
|
| 31 |
+
再统计各状态出现多少,以及它们如何相互转移。
|
| 32 |
+
矩阵记录的是转移概率,不是注意力权重。
|
| 33 |
+
|
| 34 |
+
00:01:32,600 GIS:在窗口里寻找形状
|
| 35 |
+
几何、区间和形状表征,同时读取轴间结构、局部统计和子序列匹配。
|
| 36 |
+
模板沿窗口移动,距离最低的位置给出最佳匹配。
|
| 37 |
+
保留下来的线索,包括匹配质量和出现位置。
|
| 38 |
+
|
| 39 |
+
00:01:47,433 CIS:先拼接特征
|
| 40 |
+
互补线索可以在特征层组合。
|
| 41 |
+
CIS 把卷积特征 R、区间特征 I,
|
| 42 |
+
以及频谱、时间和轴间特征 S,拼接成一个向量。
|
| 43 |
+
随后由同一个分类器读取,输出各活动的分数。
|
| 44 |
+
|
| 45 |
+
00:02:03,433 CSE / ESE:再融合分数
|
| 46 |
+
另一种组合发生在分数层:各成员先独立预测,
|
| 47 |
+
再按相同的活动顺序对齐。
|
| 48 |
+
CSE 等权平均三个成员,ESE 使用加权平均。
|
| 49 |
+
这与特征拼接,是两种不同的组合方式。
|
| 50 |
+
|
| 51 |
+
00:02:18,767 九个数据集,等权平均
|
| 52 |
+
在冻结的论文评估范围内,
|
| 53 |
+
随机搜索版 WISP 的平均 Macro F1 为百分之七十点二零。
|
| 54 |
+
十八个外部基线中,最高的 RDST 为百分之五十九点六二,
|
| 55 |
+
相差十点五八个百分点。
|
| 56 |
+
这是九个数据集的平均结果,不代表每个数据集都获胜。
|
| 57 |
+
|
| 58 |
+
00:02:37,933 增加什么信息,比增加多少更关键
|
| 59 |
+
受控消融进一步检验互补性。
|
| 60 |
+
当另外两组特征已经存在时,加入 R 和 S,
|
| 61 |
+
分别平均增加四点八零和三点零一个百分点。
|
| 62 |
+
加入 I 的平均变化为负零点二七,区间跨过零。
|
| 63 |
+
结果支持补充缺失线索,而不是不断增加重复表征。
|
| 64 |
+
|
| 65 |
+
00:02:57,767 从统一评估,到可组合表征
|
| 66 |
+
WristHARBench 提供基准数据;
|
| 67 |
+
WISP 提供方法代码、保存结果和实际保留的搜索模型。
|
| 68 |
+
三百六十个检查点不等于全部模型,也不代表完整复测。
|
| 69 |
+
从公开入口,继续探索。
|
videos/archive/20261023/WISP_zh_system_voice.mp4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:badb93c00ee41539777037c4131a5e6a2ab82822e8130795987182e26e560d95
|
| 3 |
+
size 86822326
|
videos/release_manifest.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"WISP_zh_mimo_voice.mp4": {
|
| 3 |
+
"bytes": 105702723,
|
| 4 |
+
"sha256": "a06a03c57cfc7e2cd6936fd232a7e4ee931255250eb08a4a16eff32635a1f8c1"
|
| 5 |
+
},
|
| 6 |
+
"WISP_en_captions.json": {
|
| 7 |
+
"bytes": 11864,
|
| 8 |
+
"sha256": "0f149bb3a8433feba1782b7a90517b14c2384e7048a1adf7282e52b414491409"
|
| 9 |
+
},
|
| 10 |
+
"WISP_MiMo_reproducible_project.zip": {
|
| 11 |
+
"bytes": 57091240,
|
| 12 |
+
"sha256": "797c20c808bf1db672ca298d55c4029630d9ccb16f8f8cf78e99869052349a96"
|
| 13 |
+
},
|
| 14 |
+
"WISP_en_mimo_voice.wav": {
|
| 15 |
+
"bytes": 21702444,
|
| 16 |
+
"sha256": "5707f14bbc83074c5aeea409c88faa1faed9ad6d7c804cd8857d7150c1e80081"
|
| 17 |
+
},
|
| 18 |
+
"WISP_en_cover.png": {
|
| 19 |
+
"bytes": 137541,
|
| 20 |
+
"sha256": "15d7c187ce4de39131b93cdf7b3bbea7b91cf350b2be7007d5a23d5d02ea2cfe"
|
| 21 |
+
},
|
| 22 |
+
"WISP_zh_script.txt": {
|
| 23 |
+
"bytes": 3300,
|
| 24 |
+
"sha256": "374bac0aa0158984875d3f454f6b8095dd0a3ae59d94aa250e3b56010495f2fe"
|
| 25 |
+
},
|
| 26 |
+
"WISP_zh_captions.json": {
|
| 27 |
+
"bytes": 9479,
|
| 28 |
+
"sha256": "ae495baff05ff44a11da7850fd721138f18c37836dd45454c416f2550750a03a"
|
| 29 |
+
},
|
| 30 |
+
"WISP_en_script.txt": {
|
| 31 |
+
"bytes": 3591,
|
| 32 |
+
"sha256": "70dc256ba3a5b61500090cdf7cc8dd80fac712c473c527cae3c10fc4c2bb51b7"
|
| 33 |
+
},
|
| 34 |
+
"WISP_zh_mimo_voice.wav": {
|
| 35 |
+
"bytes": 21804844,
|
| 36 |
+
"sha256": "3b9ef2e4d01bc0527ecc2ac2edfc5377c0a261b05e8d39e08ed9fd6d8e4f5726"
|
| 37 |
+
},
|
| 38 |
+
"WISP_en.vtt": {
|
| 39 |
+
"bytes": 4923,
|
| 40 |
+
"sha256": "51116e210153eed9fcbe026feb02083008c4d6fc6368f3beaf6fe51b84b38ba9"
|
| 41 |
+
},
|
| 42 |
+
"WISP_zh.srt": {
|
| 43 |
+
"bytes": 4340,
|
| 44 |
+
"sha256": "5a1bc0d4a390a592696e9a9868a6025915b81516a27ad0018b45955907243429"
|
| 45 |
+
},
|
| 46 |
+
"WISP_en.srt": {
|
| 47 |
+
"bytes": 5092,
|
| 48 |
+
"sha256": "fdc21df2bbc6cdbbae1252817ee48a36dc0c6afb1c504d9a7b2e60a2e285ed09"
|
| 49 |
+
},
|
| 50 |
+
"WISP_zh_cover.png": {
|
| 51 |
+
"bytes": 131082,
|
| 52 |
+
"sha256": "f7f33cdfcef368d2d9f484342dae5c7ab947dce60e1cf3f1890e63cab71c0775"
|
| 53 |
+
},
|
| 54 |
+
"WISP_en_mimo_voice.mp4": {
|
| 55 |
+
"bytes": 102486414,
|
| 56 |
+
"sha256": "6a84479dd6c90c65300f8a681c7dcf3a9cc2ef4cb023512f3ebf3924f3596cb1"
|
| 57 |
+
},
|
| 58 |
+
"WISP_zh.vtt": {
|
| 59 |
+
"bytes": 4216,
|
| 60 |
+
"sha256": "a1ba431ea7c28328dea30d38ae9003888be40577b8e0f730f3c1ef39a9379cb7"
|
| 61 |
+
}
|
| 62 |
+
}
|
videos/voice_release_20261028.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"date": "2026-10-28",
|
| 3 |
+
"model": "mimo-v2.5-tts",
|
| 4 |
+
"voices": {
|
| 5 |
+
"zh": "茉莉",
|
| 6 |
+
"en": "Chloe"
|
| 7 |
+
},
|
| 8 |
+
"ai_generated_narration": true,
|
| 9 |
+
"seconds": {
|
| 10 |
+
"zh": 227.13333333333333,
|
| 11 |
+
"en": 226.06666666666666
|
| 12 |
+
},
|
| 13 |
+
"caption_alignment": "Unchanged source text aligned to local Whisper word timestamps; corrected English CIS subtitle splitting.",
|
| 14 |
+
"graphics_source_date": "2026-10-18",
|
| 15 |
+
"legacy_urls": "Existing system_voice.mp4 URLs are preserved as compatibility filenames for the refreshed editions.",
|
| 16 |
+
"old_release": "Preserved in local wisp_animation_20261018 and WISP/videos/archive/20261023."
|
| 17 |
+
}
|