chenmohan commited on
Commit
3e6d980
·
verified ·
1 Parent(s): 9d291ae

发布 TinyPose 0.2.0 weights

Browse files
README.md CHANGED
@@ -1,19 +1,20 @@
1
- ---
2
- license: apache-2.0
3
- pipeline_tag: keypoint-detection
4
- tags:
5
- - onnx
6
- - webgpu
7
- - wasm
8
- ---
9
-
10
- # PP-TinyPose Web SDK 模型
11
-
12
- [中文模型卡](tinypose-256x192/0.1.0/README.md) · [English model card](tinypose-256x192/0.1.0/README.en.md)
13
-
14
- 官方增强版 256×192 FP32,版本 0.1.0,5685847 字节。
15
- 仅提供单人图或外部人体框的 17 点姿态估计。桌面 CPU/GPU 已有验收;不声明手机或 NPU 兼容。
16
- The enhanced 256×192 FP32 model estimates 17 keypoints for one person or a caller-provided crop. Desktop CPU/GPU evidence is available; mobile and NPU are unverified.
17
-
18
- 上游仓库 Apache-2.0;ZIP 没有独立权重许可。完整来源、限制、许可及摘要见模型卡。
19
- Upstream uses Apache-2.0; the ZIP has no separate weight license. See the model card for provenance, limitations and hashes.
 
 
1
+ ---
2
+ license: apache-2.0
3
+ pipeline_tag: keypoint-detection
4
+ tags:
5
+ - onnx
6
+ - webgpu
7
+ - wasm
8
+ ---
9
+
10
+ # PP-TinyPose Web SDK 模型
11
+
12
+ 稳定清单包含三个通过项:
13
+
14
+ - [256×192 FP32 0.1.0](tinypose-256x192/0.1.0/README.md)(默认)
15
+ - [128×96 FP32 0.2.0](tinypose-128x96/0.2.0/fp32/README.md)
16
+ - [128×96 FP16 权重(FP32 计算)0.2.0](tinypose-128x96/0.2.0/w16a32/README.md)
17
+
18
+ English cards are linked from each entry. All three passed the dated desktop WASM/WebGPU × main/worker checks. The 64-image GT-box subset reports mean OKS rather than full COCO AP. Mobile and NPU remain unverified.
19
+
20
+ 上游仓库采用 Apache-2.0;官方 ZIP 没有独立权重许可。W16A32 仅压缩权重存储,不声明 FP16 计算或加速。完整来源、修改、限制、许可与摘要见各模型卡。
tinypose-128x96/0.2.0/fp32/LICENSE ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Apache License
2
+ Version 2.0, January 2004
3
+ http://www.apache.org/licenses/
4
+
5
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
6
+
7
+ 1. Definitions.
8
+
9
+ "License" shall mean the terms and conditions for use, reproduction,
10
+ and distribution as defined by Sections 1 through 9 of this document.
11
+
12
+ "Licensor" shall mean the copyright owner or entity authorized by
13
+ the copyright owner that is granting the License.
14
+
15
+ "Legal Entity" shall mean the union of the acting entity and all
16
+ other entities that control, are controlled by, or are under common
17
+ control with that entity. For the purposes of this definition,
18
+ "control" means (i) the power, direct or indirect, to cause the
19
+ direction or management of such entity, whether by contract or
20
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
21
+ outstanding shares, or (iii) beneficial ownership of such entity.
22
+
23
+ "You" (or "Your") shall mean an individual or Legal Entity
24
+ exercising permissions granted by this License.
25
+
26
+ "Source" form shall mean the preferred form for making modifications,
27
+ including but not limited to software source code, documentation
28
+ source, and configuration files.
29
+
30
+ "Object" form shall mean any form resulting from mechanical
31
+ transformation or translation of a Source form, including but
32
+ not limited to compiled object code, generated documentation,
33
+ and conversions to other media types.
34
+
35
+ "Work" shall mean the work of authorship, whether in Source or
36
+ Object form, made available under the License, as indicated by a
37
+ copyright notice that is included in or attached to the work
38
+ (an example is provided in the Appendix below).
39
+
40
+ "Derivative Works" shall mean any work, whether in Source or Object
41
+ form, that is based on (or derived from) the Work and for which the
42
+ editorial revisions, annotations, elaborations, or other modifications
43
+ represent, as a whole, an original work of authorship. For the purposes
44
+ of this License, Derivative Works shall not include works that remain
45
+ separable from, or merely link (or bind by name) to the interfaces of,
46
+ the Work and Derivative Works thereof.
47
+
48
+ "Contribution" shall mean any work of authorship, including
49
+ the original version of the Work and any modifications or additions
50
+ to that Work or Derivative Works thereof, that is intentionally
51
+ submitted to Licensor for inclusion in the Work by the copyright owner
52
+ or by an individual or Legal Entity authorized to submit on behalf of
53
+ the copyright owner. For the purposes of this definition, "submitted"
54
+ means any form of electronic, verbal, or written communication sent
55
+ to the Licensor or its representatives, including but not limited to
56
+ communication on electronic mailing lists, source code control systems,
57
+ and issue tracking systems that are managed by, or on behalf of, the
58
+ Licensor for the purpose of discussing and improving the Work, but
59
+ excluding communication that is conspicuously marked or otherwise
60
+ designated in writing by the copyright owner as "Not a Contribution."
61
+
62
+ "Contributor" shall mean Licensor and any individual or Legal Entity
63
+ on behalf of whom a Contribution has been received by Licensor and
64
+ subsequently incorporated within the Work.
65
+
66
+ 2. Grant of Copyright License. Subject to the terms and conditions of
67
+ this License, each Contributor hereby grants to You a perpetual,
68
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
69
+ copyright license to reproduce, prepare Derivative Works of,
70
+ publicly display, publicly perform, sublicense, and distribute the
71
+ Work and such Derivative Works in Source or Object form.
72
+
73
+ 3. Grant of Patent License. Subject to the terms and conditions of
74
+ this License, each Contributor hereby grants to You a perpetual,
75
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
76
+ (except as stated in this section) patent license to make, have made,
77
+ use, offer to sell, sell, import, and otherwise transfer the Work,
78
+ where such license applies only to those patent claims licensable
79
+ by such Contributor that are necessarily infringed by their
80
+ Contribution(s) alone or by combination of their Contribution(s)
81
+ with the Work to which such Contribution(s) was submitted. If You
82
+ institute patent litigation against any entity (including a
83
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
84
+ or a Contribution incorporated within the Work constitutes direct
85
+ or contributory patent infringement, then any patent licenses
86
+ granted to You under this License for that Work shall terminate
87
+ as of the date such litigation is filed.
88
+
89
+ 4. Redistribution. You may reproduce and distribute copies of the
90
+ Work or Derivative Works thereof in any medium, with or without
91
+ modifications, and in Source or Object form, provided that You
92
+ meet the following conditions:
93
+
94
+ (a) You must give any other recipients of the Work or
95
+ Derivative Works a copy of this License; and
96
+
97
+ (b) You must cause any modified files to carry prominent notices
98
+ stating that You changed the files; and
99
+
100
+ (c) You must retain, in the Source form of any Derivative Works
101
+ that You distribute, all copyright, patent, trademark, and
102
+ attribution notices from the Source form of the Work,
103
+ excluding those notices that do not pertain to any part of
104
+ the Derivative Works; and
105
+
106
+ (d) If the Work includes a "NOTICE" text file as part of its
107
+ distribution, then any Derivative Works that You distribute must
108
+ include a readable copy of the attribution notices contained
109
+ within such NOTICE file, excluding those notices that do not
110
+ pertain to any part of the Derivative Works, in at least one
111
+ of the following places: within a NOTICE text file distributed
112
+ as part of the Derivative Works; within the Source form or
113
+ documentation, if provided along with the Derivative Works; or,
114
+ within a display generated by the Derivative Works, if and
115
+ wherever such third-party notices normally appear. The contents
116
+ of the NOTICE file are for informational purposes only and
117
+ do not modify the License. You may add Your own attribution
118
+ notices within Derivative Works that You distribute, alongside
119
+ or as an addendum to the NOTICE text from the Work, provided
120
+ that such additional attribution notices cannot be construed
121
+ as modifying the License.
122
+
123
+ You may add Your own copyright statement to Your modifications and
124
+ may provide additional or different license terms and conditions
125
+ for use, reproduction, or distribution of Your modifications, or
126
+ for any such Derivative Works as a whole, provided Your use,
127
+ reproduction, and distribution of the Work otherwise complies with
128
+ the conditions stated in this License.
129
+
130
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
131
+ any Contribution intentionally submitted for inclusion in the Work
132
+ by You to the Licensor shall be under the terms and conditions of
133
+ this License, without any additional terms or conditions.
134
+ Notwithstanding the above, nothing herein shall supersede or modify
135
+ the terms of any separate license agreement you may have executed
136
+ with Licensor regarding such Contributions.
137
+
138
+ 6. Trademarks. This License does not grant permission to use the trade
139
+ names, trademarks, service marks, or product names of the Licensor,
140
+ except as required for reasonable and customary use in describing the
141
+ origin of the Work and reproducing the content of the NOTICE file.
142
+
143
+ 7. Disclaimer of Warranty. Unless required by applicable law or
144
+ agreed to in writing, Licensor provides the Work (and each
145
+ Contributor provides its Contributions) on an "AS IS" BASIS,
146
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147
+ implied, including, without limitation, any warranties or conditions
148
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149
+ PARTICULAR PURPOSE. You are solely responsible for determining the
150
+ appropriateness of using or redistributing the Work and assume any
151
+ risks associated with Your exercise of permissions under this License.
152
+
153
+ 8. Limitation of Liability. In no event and under no legal theory,
154
+ whether in tort (including negligence), contract, or otherwise,
155
+ unless required by applicable law (such as deliberate and grossly
156
+ negligent acts) or agreed to in writing, shall any Contributor be
157
+ liable to You for damages, including any direct, indirect, special,
158
+ incidental, or consequential damages of any character arising as a
159
+ result of this License or out of the use or inability to use the
160
+ Work (including but not limited to damages for loss of goodwill,
161
+ work stoppage, computer failure or malfunction, or any and all
162
+ other commercial damages or losses), even if such Contributor
163
+ has been advised of the possibility of such damages.
164
+
165
+ 9. Accepting Warranty or Additional Liability. While redistributing
166
+ the Work or Derivative Works thereof, You may choose to offer,
167
+ and charge a fee for, acceptance of support, warranty, indemnity,
168
+ or other liability obligations and/or rights consistent with this
169
+ License. However, in accepting such obligations, You may act only
170
+ on Your own behalf and on Your sole responsibility, not on behalf
171
+ of any other Contributor, and only if You agree to indemnify,
172
+ defend, and hold each Contributor harmless for any liability
173
+ incurred by, or claims asserted against, such Contributor by reason
174
+ of your accepting any such warranty or additional liability.
175
+
176
+ END OF TERMS AND CONDITIONS
177
+
178
+ APPENDIX: How to apply the Apache License to your work.
179
+
180
+ To apply the Apache License to your work, attach the following
181
+ boilerplate notice, with the fields enclosed by brackets "[]"
182
+ replaced with your own identifying information. (Don't include
183
+ the brackets!) The text should be enclosed in the appropriate
184
+ comment syntax for the file format. We also recommend that a
185
+ file or class name and description of purpose be included on the
186
+ same "printed page" as the copyright notice for easier
187
+ identification within third-party archives.
188
+
189
+ Copyright [yyyy] [name of copyright owner]
190
+
191
+ Licensed under the Apache License, Version 2.0 (the "License");
192
+ you may not use this file except in compliance with the License.
193
+ You may obtain a copy of the License at
194
+
195
+ http://www.apache.org/licenses/LICENSE-2.0
196
+
197
+ Unless required by applicable law or agreed to in writing, software
198
+ distributed under the License is distributed on an "AS IS" BASIS,
199
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200
+ See the License for the specific language governing permissions and
201
+ limitations under the License.
tinypose-128x96/0.2.0/fp32/NOTICE ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ PP-TinyPose 浏览器 SDK
2
+
3
+ 模型与参考算法来源:PaddlePaddle/PaddleDetection。
4
+ Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
5
+ Apache License, Version 2.0,详见 LICENSE。
6
+ 固定源码:b25522a0f4bde8c80603f3ba5e3472059972e3b5。
7
+
8
+ 裁剪、TopDownEvalAffine 与 DARK 的 TypeScript 实现基于以下文件改编:
9
+ deploy/python/keypoint_preprocess.py
10
+ deploy/python/keypoint_postprocess.py
11
+ deploy/python/keypoint_infer.py
12
+ 原预处理引用 OpenMMLab/mmpose 的 post_transforms.py;DARK 算法引用
13
+ ilovepose/DarkPose 的 lib/core/inference.py。本仓库保留上游说明,未复制训练代码。
14
+
15
+ 权重来自官方 tinypose_enhance/tinypose_256x192.zip 与
16
+ tinypose_enhance/tinypose_128x96.zip 模型表。
17
+ ZIP 不包含独立权重许可证,发布归属与仓库 Apache-2.0 许可一并记录;
18
+ COCO、AI Challenger 和上游业务训练数据的说明不构成额外的权重授权书。
19
+
20
+ 128×96 W16A32 版本由本仓库修改:270 个直接供给 Conv 的常量权重以
21
+ float16 保存,并在算子前显式 Cast 回 float32;激活、算子和输入输出保持
22
+ float32。该版本是存储压缩,不声明 FP16 计算。
23
+
24
+ Demo 样图来自上述固定源码 demo/hrnet_demo.jpg,保留为上游姿态示例。
25
+ 运行依赖 ONNX Runtime Web(Microsoft,MIT);其发行包保留自身许可。
26
+ 共享 UI token 来源:chenmohan123.github.io/standards/v1/ui-tokens.json。
27
+ Demo 图标使用 Lucide(ISC);仅用于 Demo,不是 SDK 运行依赖。
tinypose-128x96/0.2.0/fp32/README.en.md ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # PP-TinyPose 128×96 FP32
2
+
3
+ [中文](README.md)
4
+
5
+ This mirror is maintained by chenmohan, not the official PaddleDetection account. It contains the official enhanced TinyPose model.
6
+ Version 0.2.0; 5685846 bytes; SHA-256 `a0e2edd5272f48243a9cbd571151eda966f1bfa865a954f39e1e344aa5a14cf8`; ONNX opset 17. The official report states about 1.32M parameters; distribution metadata leaves the exact count unset because it has not been independently established.
7
+
8
+ Pinned upstream commit `b25522a0f4bde8c80603f3ba5e3472059972e3b5`. Official deployment archive: https://bj.bcebos.com/v1/paddledet/models/keypoint/tinypose_enhance/tinypose_128x96.zip .
9
+ The upstream repository uses Apache-2.0 (see LICENSE and NOTICE). The ZIP contains no separate weight license. This mirror records the official provenance and repository license without inferring additional permission.
10
+ The converted graph is unchanged. See conversion.json for pinned ZIP, internal file, ONNX and tool hashes.
11
+
12
+ Input is one person as RGB float32 NCHW `[1,3,128,96]`; outputs are `[1,17,32,24]` heatmaps and argmax. The SDK applies upstream crop expansion, TopDownEvalAffine and DARK to return 17 COCO keypoints in original-image coordinates. Scores are uncalibrated heatmap responses, not visibility probabilities.
13
+
14
+ On the fixed 64-image, 110-person GT-box subset dated 2026-09-17, mean OKS was `0.727050`; this is not full COCO AP. WASM/WebGPU × main/worker passed on desktop Chromium 153, Windows 11 and ORT Web 1.27.0. W16A32 reduced model bytes by 44.584% against same-size FP32; it does not claim faster execution or FP16 compute. Timings are observations from the fixed device only.
15
+
16
+ Automatic multi-person detection, video tracking and NPU are outside scope; mobile is unverified. SDK, reproduction and evidence: https://github.com/chenmohan123/web-sdk-PP-TinyPose
17
+ ModelScope and Hugging Face use immutable revisions. ModelScope is the default; an explicit source failure never switches sources. npm contains no ONNX weights.
tinypose-128x96/0.2.0/fp32/README.md ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # PP-TinyPose 128×96 FP32
2
+
3
+ [English](README.en.md)
4
+
5
+ 本镜像由 chenmohan 维护,来源于 PaddleDetection 官方增强版 TinyPose;并非官方账号。
6
+ 版本 0.2.0,模型 5685846 字节,SHA-256 `a0e2edd5272f48243a9cbd571151eda966f1bfa865a954f39e1e344aa5a14cf8`,ONNX opset 17;官方报告约 1.32M 参数,当前分发清单不填未经独立证明的精确参数量。
7
+
8
+ 上游固定提交 `b25522a0f4bde8c80603f3ba5e3472059972e3b5`;官方部署 ZIP:https://bj.bcebos.com/v1/paddledet/models/keypoint/tinypose_enhance/tinypose_128x96.zip 。
9
+ 源码为 Apache-2.0,完整原文见 LICENSE,归因见 NOTICE。官方 ZIP 没有独立权重许可证;本镜像如实记录官方模型发布归属及仓库许可,不推断额外授权。
10
+ 转换图未修改。 固定 ZIP、内部文件、ONNX 和工具摘要见 conversion.json。
11
+
12
+ 输入单个人体 RGB float32 NCHW `[1,3,128,96]`;输出 `[1,17,32,24]` 热力图与 argmax。SDK 使用上游扩框、TopDownEvalAffine 和 DARK 还原原图 COCO 17 点。score 是未校准热力图响应,不是可见性概率。
13
+
14
+ 2026-09-17 固定 64 图、110 人 GT 框子集平均 OKS 为 `0.727050`,不是 COCO 全量 AP。桌面 Chromium 153、Windows 11、ORT Web 1.27.0 的 WASM/WebGPU × main/worker 已通过。W16A32 相对同规格 FP32 体积减少 44.584%,不宣称加速或 FP16 计算。性能仅是固定设备观测。
15
+
16
+ 不包含自动多人检测、视频跟踪或 NPU;手机未验收。SDK、重建脚本与证据:https://github.com/chenmohan123/web-sdk-PP-TinyPose
17
+ 模型按固定提交从 ModelScope/Hugging Face 分发,默认 ModelScope;显式来源失败不静默换源。npm 不包含 ONNX。
tinypose-128x96/0.2.0/fp32/conversion.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "0.2.0",
3
+ "upstreamRevision": "b25522a0f4bde8c80603f3ba5e3472059972e3b5",
4
+ "archive": {
5
+ "url": "https://bj.bcebos.com/v1/paddledet/models/keypoint/tinypose_enhance/tinypose_128x96.zip",
6
+ "path": "F:\\git\\00_chenmohan\\github\\web-sdk-PP-TinyPose\\.worktrees\\tinypose-variants\\.tmp\\variants\\downloads\\tinypose_128x96.zip",
7
+ "bytes": 5397578,
8
+ "sha256": "b737aa851a1ce9403950280be5df4bdd9c0e0287dd7ecdaa16f534d9c940015d"
9
+ },
10
+ "paddleAssets": [
11
+ {
12
+ "path": "infer_cfg.yml",
13
+ "bytes": 343,
14
+ "sha256": "b3f7e023bfd403144455d75bfd6028cec6c5283d4ad8398cf24135948f8b7720"
15
+ },
16
+ {
17
+ "path": "model.pdiparams.info",
18
+ "bytes": 135248,
19
+ "sha256": "62b1cb256f4c53100e893ed4c07023506ac4e834c86bcf8a343f35a4280f3480"
20
+ },
21
+ {
22
+ "path": "model.pdiparams",
23
+ "bytes": 5456137,
24
+ "sha256": "8ae7262a63a27d7011bd9ac7993bd261ba1c5dbdfae1b2753dca9711b0b4b665"
25
+ },
26
+ {
27
+ "path": "model.pdmodel",
28
+ "bytes": 10928441,
29
+ "sha256": "4a416c423b2ddb8ca0e76b8edd965051cf35981a3ac331f6a1952217e97056bb"
30
+ }
31
+ ],
32
+ "model": {
33
+ "bytes": 5685846,
34
+ "sha256": "a0e2edd5272f48243a9cbd571151eda966f1bfa865a954f39e1e344aa5a14cf8"
35
+ },
36
+ "opset": 17,
37
+ "input": [
38
+ 1,
39
+ 3,
40
+ 128,
41
+ 96
42
+ ],
43
+ "output": [
44
+ 1,
45
+ 17,
46
+ 32,
47
+ 24
48
+ ],
49
+ "parameterCount": null,
50
+ "tools": {
51
+ "python": "3.11.15 (main, Jul 18 2026, 17:06:27) [MSC v.1944 64 bit (AMD64)]",
52
+ "os": "Windows-10-10.0.26200-SP0",
53
+ "paddle": "2.6.2",
54
+ "paddle2onnx": "1.3.1",
55
+ "onnx": "1.16.2",
56
+ "onnxruntime": "1.20.1",
57
+ "onnxconverterCommon": "1.16.0",
58
+ "numpy": "1.26.4",
59
+ "opencv": "4.11.0"
60
+ },
61
+ "reproduction": "tools/variants/prepare.py",
62
+ "evidence": "reports/2026-09-17-variants",
63
+ "graphModification": "无;Paddle2ONNX 1.3.1 opset 17 原样转换"
64
+ }
tinypose-128x96/0.2.0/fp32/tinypose-128x96-fp32.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0e2edd5272f48243a9cbd571151eda966f1bfa865a954f39e1e344aa5a14cf8
3
+ size 5685846
tinypose-128x96/0.2.0/w16a32/LICENSE ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Apache License
2
+ Version 2.0, January 2004
3
+ http://www.apache.org/licenses/
4
+
5
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
6
+
7
+ 1. Definitions.
8
+
9
+ "License" shall mean the terms and conditions for use, reproduction,
10
+ and distribution as defined by Sections 1 through 9 of this document.
11
+
12
+ "Licensor" shall mean the copyright owner or entity authorized by
13
+ the copyright owner that is granting the License.
14
+
15
+ "Legal Entity" shall mean the union of the acting entity and all
16
+ other entities that control, are controlled by, or are under common
17
+ control with that entity. For the purposes of this definition,
18
+ "control" means (i) the power, direct or indirect, to cause the
19
+ direction or management of such entity, whether by contract or
20
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
21
+ outstanding shares, or (iii) beneficial ownership of such entity.
22
+
23
+ "You" (or "Your") shall mean an individual or Legal Entity
24
+ exercising permissions granted by this License.
25
+
26
+ "Source" form shall mean the preferred form for making modifications,
27
+ including but not limited to software source code, documentation
28
+ source, and configuration files.
29
+
30
+ "Object" form shall mean any form resulting from mechanical
31
+ transformation or translation of a Source form, including but
32
+ not limited to compiled object code, generated documentation,
33
+ and conversions to other media types.
34
+
35
+ "Work" shall mean the work of authorship, whether in Source or
36
+ Object form, made available under the License, as indicated by a
37
+ copyright notice that is included in or attached to the work
38
+ (an example is provided in the Appendix below).
39
+
40
+ "Derivative Works" shall mean any work, whether in Source or Object
41
+ form, that is based on (or derived from) the Work and for which the
42
+ editorial revisions, annotations, elaborations, or other modifications
43
+ represent, as a whole, an original work of authorship. For the purposes
44
+ of this License, Derivative Works shall not include works that remain
45
+ separable from, or merely link (or bind by name) to the interfaces of,
46
+ the Work and Derivative Works thereof.
47
+
48
+ "Contribution" shall mean any work of authorship, including
49
+ the original version of the Work and any modifications or additions
50
+ to that Work or Derivative Works thereof, that is intentionally
51
+ submitted to Licensor for inclusion in the Work by the copyright owner
52
+ or by an individual or Legal Entity authorized to submit on behalf of
53
+ the copyright owner. For the purposes of this definition, "submitted"
54
+ means any form of electronic, verbal, or written communication sent
55
+ to the Licensor or its representatives, including but not limited to
56
+ communication on electronic mailing lists, source code control systems,
57
+ and issue tracking systems that are managed by, or on behalf of, the
58
+ Licensor for the purpose of discussing and improving the Work, but
59
+ excluding communication that is conspicuously marked or otherwise
60
+ designated in writing by the copyright owner as "Not a Contribution."
61
+
62
+ "Contributor" shall mean Licensor and any individual or Legal Entity
63
+ on behalf of whom a Contribution has been received by Licensor and
64
+ subsequently incorporated within the Work.
65
+
66
+ 2. Grant of Copyright License. Subject to the terms and conditions of
67
+ this License, each Contributor hereby grants to You a perpetual,
68
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
69
+ copyright license to reproduce, prepare Derivative Works of,
70
+ publicly display, publicly perform, sublicense, and distribute the
71
+ Work and such Derivative Works in Source or Object form.
72
+
73
+ 3. Grant of Patent License. Subject to the terms and conditions of
74
+ this License, each Contributor hereby grants to You a perpetual,
75
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
76
+ (except as stated in this section) patent license to make, have made,
77
+ use, offer to sell, sell, import, and otherwise transfer the Work,
78
+ where such license applies only to those patent claims licensable
79
+ by such Contributor that are necessarily infringed by their
80
+ Contribution(s) alone or by combination of their Contribution(s)
81
+ with the Work to which such Contribution(s) was submitted. If You
82
+ institute patent litigation against any entity (including a
83
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
84
+ or a Contribution incorporated within the Work constitutes direct
85
+ or contributory patent infringement, then any patent licenses
86
+ granted to You under this License for that Work shall terminate
87
+ as of the date such litigation is filed.
88
+
89
+ 4. Redistribution. You may reproduce and distribute copies of the
90
+ Work or Derivative Works thereof in any medium, with or without
91
+ modifications, and in Source or Object form, provided that You
92
+ meet the following conditions:
93
+
94
+ (a) You must give any other recipients of the Work or
95
+ Derivative Works a copy of this License; and
96
+
97
+ (b) You must cause any modified files to carry prominent notices
98
+ stating that You changed the files; and
99
+
100
+ (c) You must retain, in the Source form of any Derivative Works
101
+ that You distribute, all copyright, patent, trademark, and
102
+ attribution notices from the Source form of the Work,
103
+ excluding those notices that do not pertain to any part of
104
+ the Derivative Works; and
105
+
106
+ (d) If the Work includes a "NOTICE" text file as part of its
107
+ distribution, then any Derivative Works that You distribute must
108
+ include a readable copy of the attribution notices contained
109
+ within such NOTICE file, excluding those notices that do not
110
+ pertain to any part of the Derivative Works, in at least one
111
+ of the following places: within a NOTICE text file distributed
112
+ as part of the Derivative Works; within the Source form or
113
+ documentation, if provided along with the Derivative Works; or,
114
+ within a display generated by the Derivative Works, if and
115
+ wherever such third-party notices normally appear. The contents
116
+ of the NOTICE file are for informational purposes only and
117
+ do not modify the License. You may add Your own attribution
118
+ notices within Derivative Works that You distribute, alongside
119
+ or as an addendum to the NOTICE text from the Work, provided
120
+ that such additional attribution notices cannot be construed
121
+ as modifying the License.
122
+
123
+ You may add Your own copyright statement to Your modifications and
124
+ may provide additional or different license terms and conditions
125
+ for use, reproduction, or distribution of Your modifications, or
126
+ for any such Derivative Works as a whole, provided Your use,
127
+ reproduction, and distribution of the Work otherwise complies with
128
+ the conditions stated in this License.
129
+
130
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
131
+ any Contribution intentionally submitted for inclusion in the Work
132
+ by You to the Licensor shall be under the terms and conditions of
133
+ this License, without any additional terms or conditions.
134
+ Notwithstanding the above, nothing herein shall supersede or modify
135
+ the terms of any separate license agreement you may have executed
136
+ with Licensor regarding such Contributions.
137
+
138
+ 6. Trademarks. This License does not grant permission to use the trade
139
+ names, trademarks, service marks, or product names of the Licensor,
140
+ except as required for reasonable and customary use in describing the
141
+ origin of the Work and reproducing the content of the NOTICE file.
142
+
143
+ 7. Disclaimer of Warranty. Unless required by applicable law or
144
+ agreed to in writing, Licensor provides the Work (and each
145
+ Contributor provides its Contributions) on an "AS IS" BASIS,
146
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147
+ implied, including, without limitation, any warranties or conditions
148
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149
+ PARTICULAR PURPOSE. You are solely responsible for determining the
150
+ appropriateness of using or redistributing the Work and assume any
151
+ risks associated with Your exercise of permissions under this License.
152
+
153
+ 8. Limitation of Liability. In no event and under no legal theory,
154
+ whether in tort (including negligence), contract, or otherwise,
155
+ unless required by applicable law (such as deliberate and grossly
156
+ negligent acts) or agreed to in writing, shall any Contributor be
157
+ liable to You for damages, including any direct, indirect, special,
158
+ incidental, or consequential damages of any character arising as a
159
+ result of this License or out of the use or inability to use the
160
+ Work (including but not limited to damages for loss of goodwill,
161
+ work stoppage, computer failure or malfunction, or any and all
162
+ other commercial damages or losses), even if such Contributor
163
+ has been advised of the possibility of such damages.
164
+
165
+ 9. Accepting Warranty or Additional Liability. While redistributing
166
+ the Work or Derivative Works thereof, You may choose to offer,
167
+ and charge a fee for, acceptance of support, warranty, indemnity,
168
+ or other liability obligations and/or rights consistent with this
169
+ License. However, in accepting such obligations, You may act only
170
+ on Your own behalf and on Your sole responsibility, not on behalf
171
+ of any other Contributor, and only if You agree to indemnify,
172
+ defend, and hold each Contributor harmless for any liability
173
+ incurred by, or claims asserted against, such Contributor by reason
174
+ of your accepting any such warranty or additional liability.
175
+
176
+ END OF TERMS AND CONDITIONS
177
+
178
+ APPENDIX: How to apply the Apache License to your work.
179
+
180
+ To apply the Apache License to your work, attach the following
181
+ boilerplate notice, with the fields enclosed by brackets "[]"
182
+ replaced with your own identifying information. (Don't include
183
+ the brackets!) The text should be enclosed in the appropriate
184
+ comment syntax for the file format. We also recommend that a
185
+ file or class name and description of purpose be included on the
186
+ same "printed page" as the copyright notice for easier
187
+ identification within third-party archives.
188
+
189
+ Copyright [yyyy] [name of copyright owner]
190
+
191
+ Licensed under the Apache License, Version 2.0 (the "License");
192
+ you may not use this file except in compliance with the License.
193
+ You may obtain a copy of the License at
194
+
195
+ http://www.apache.org/licenses/LICENSE-2.0
196
+
197
+ Unless required by applicable law or agreed to in writing, software
198
+ distributed under the License is distributed on an "AS IS" BASIS,
199
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200
+ See the License for the specific language governing permissions and
201
+ limitations under the License.
tinypose-128x96/0.2.0/w16a32/NOTICE ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ PP-TinyPose 浏览器 SDK
2
+
3
+ 模型与参考算法来源:PaddlePaddle/PaddleDetection。
4
+ Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
5
+ Apache License, Version 2.0,详见 LICENSE。
6
+ 固定源码:b25522a0f4bde8c80603f3ba5e3472059972e3b5。
7
+
8
+ 裁剪、TopDownEvalAffine 与 DARK 的 TypeScript 实现基于以下文件改编:
9
+ deploy/python/keypoint_preprocess.py
10
+ deploy/python/keypoint_postprocess.py
11
+ deploy/python/keypoint_infer.py
12
+ 原预处理引用 OpenMMLab/mmpose 的 post_transforms.py;DARK 算法引用
13
+ ilovepose/DarkPose 的 lib/core/inference.py。本仓库保留上游说明,未复制训练代码。
14
+
15
+ 权重来自官方 tinypose_enhance/tinypose_256x192.zip 与
16
+ tinypose_enhance/tinypose_128x96.zip 模型表。
17
+ ZIP 不包含独立权重许可证,发布归属与仓库 Apache-2.0 许可一并记录;
18
+ COCO、AI Challenger 和上游业务训练数据的说明不构成额外的权重授权书。
19
+
20
+ 128×96 W16A32 版本由本仓库修改:270 个直接供给 Conv 的常量权重以
21
+ float16 保存,并在算子前显式 Cast 回 float32;激活、算子和输入输出保持
22
+ float32。该版本是存储压缩,不声明 FP16 计算。
23
+
24
+ Demo 样图来自上述固定源码 demo/hrnet_demo.jpg,保留为上游姿态示例。
25
+ 运行依赖 ONNX Runtime Web(Microsoft,MIT);其发行包保留自身许可。
26
+ 共享 UI token 来源:chenmohan123.github.io/standards/v1/ui-tokens.json。
27
+ Demo 图标使用 Lucide(ISC);仅用于 Demo,不是 SDK 运行依赖。
tinypose-128x96/0.2.0/w16a32/README.en.md ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # PP-TinyPose 128×96 FP16 weights (FP32 compute)
2
+
3
+ [中文](README.md)
4
+
5
+ This mirror is maintained by chenmohan, not the official PaddleDetection account. It contains the official enhanced TinyPose model.
6
+ Version 0.2.0; 3150847 bytes; SHA-256 `8671c7b424d85d4f017b6410602de6095b1fbb9fffca38ad5489039dd2a0cfea`; ONNX opset 17. The official report states about 1.32M parameters; distribution metadata leaves the exact count unset because it has not been independently established.
7
+
8
+ Pinned upstream commit `b25522a0f4bde8c80603f3ba5e3472059972e3b5`. Official deployment archive: https://bj.bcebos.com/v1/paddledet/models/keypoint/tinypose_enhance/tinypose_128x96.zip .
9
+ The upstream repository uses Apache-2.0 (see LICENSE and NOTICE). The ZIP contains no separate weight license. This mirror records the official provenance and repository license without inferring additional permission.
10
+ 270 constant weights consumed by Conv are stored as FP16 and explicitly cast back to FP32 before Conv; activations, operators, inputs and outputs remain FP32. See conversion.json for pinned ZIP, internal file, ONNX and tool hashes.
11
+
12
+ Input is one person as RGB float32 NCHW `[1,3,128,96]`; outputs are `[1,17,32,24]` heatmaps and argmax. The SDK applies upstream crop expansion, TopDownEvalAffine and DARK to return 17 COCO keypoints in original-image coordinates. Scores are uncalibrated heatmap responses, not visibility probabilities.
13
+
14
+ On the fixed 64-image, 110-person GT-box subset dated 2026-09-17, mean OKS was `0.727106`; this is not full COCO AP. WASM/WebGPU × main/worker passed on desktop Chromium 153, Windows 11 and ORT Web 1.27.0. W16A32 reduced model bytes by 44.584% against same-size FP32; it does not claim faster execution or FP16 compute. Timings are observations from the fixed device only.
15
+
16
+ Automatic multi-person detection, video tracking and NPU are outside scope; mobile is unverified. SDK, reproduction and evidence: https://github.com/chenmohan123/web-sdk-PP-TinyPose
17
+ ModelScope and Hugging Face use immutable revisions. ModelScope is the default; an explicit source failure never switches sources. npm contains no ONNX weights.
tinypose-128x96/0.2.0/w16a32/README.md ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # PP-TinyPose 128×96 FP16 权重(FP32 计算)
2
+
3
+ [English](README.en.md)
4
+
5
+ 本镜像由 chenmohan 维护,来源于 PaddleDetection 官方增强版 TinyPose;并非官方账号。
6
+ 版本 0.2.0,模型 3150847 字节,SHA-256 `8671c7b424d85d4f017b6410602de6095b1fbb9fffca38ad5489039dd2a0cfea`,ONNX opset 17;官方报告约 1.32M 参数,当前分发清单不填未经独立证明的精确参数量。
7
+
8
+ 上游固定提交 `b25522a0f4bde8c80603f3ba5e3472059972e3b5`;官方部署 ZIP:https://bj.bcebos.com/v1/paddledet/models/keypoint/tinypose_enhance/tinypose_128x96.zip 。
9
+ 源码为 Apache-2.0,完整原文见 LICENSE,归因见 NOTICE。官方 ZIP 没有独立权重许可证;本镜像如实记录官方模型发布归属及仓库许可,不推断额外授权。
10
+ 270 个直接供给 Conv 的常量权重以 FP16 保存,并在 Conv 前显式 Cast 回 FP32;激活、算子和输入输出均为 FP32。 固定 ZIP、内部文件、ONNX 和工具摘要见 conversion.json。
11
+
12
+ 输入单个人体 RGB float32 NCHW `[1,3,128,96]`;输出 `[1,17,32,24]` 热力图与 argmax。SDK 使用上游扩框、TopDownEvalAffine 和 DARK 还原原图 COCO 17 点。score 是未校准热力图响应,不是可见性概率。
13
+
14
+ 2026-09-17 固定 64 图、110 人 GT 框子集平均 OKS 为 `0.727106`,不是 COCO 全量 AP。桌面 Chromium 153、Windows 11、ORT Web 1.27.0 的 WASM/WebGPU × main/worker 已通过。W16A32 相对同规格 FP32 体积减少 44.584%,不宣称加速或 FP16 计算。性能仅是固定设备观测。
15
+
16
+ 不包含自动多人检测、视频跟踪或 NPU;手机未验收。SDK、重建脚本与证据:https://github.com/chenmohan123/web-sdk-PP-TinyPose
17
+ 模型按固定提交从 ModelScope/Hugging Face 分发,默认 ModelScope;显式来源失败不静默换源。npm 不包含 ONNX。
tinypose-128x96/0.2.0/w16a32/conversion.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "0.2.0",
3
+ "upstreamRevision": "b25522a0f4bde8c80603f3ba5e3472059972e3b5",
4
+ "archive": {
5
+ "url": "https://bj.bcebos.com/v1/paddledet/models/keypoint/tinypose_enhance/tinypose_128x96.zip",
6
+ "path": "F:\\git\\00_chenmohan\\github\\web-sdk-PP-TinyPose\\.worktrees\\tinypose-variants\\.tmp\\variants\\downloads\\tinypose_128x96.zip",
7
+ "bytes": 5397578,
8
+ "sha256": "b737aa851a1ce9403950280be5df4bdd9c0e0287dd7ecdaa16f534d9c940015d"
9
+ },
10
+ "paddleAssets": [
11
+ {
12
+ "path": "infer_cfg.yml",
13
+ "bytes": 343,
14
+ "sha256": "b3f7e023bfd403144455d75bfd6028cec6c5283d4ad8398cf24135948f8b7720"
15
+ },
16
+ {
17
+ "path": "model.pdiparams.info",
18
+ "bytes": 135248,
19
+ "sha256": "62b1cb256f4c53100e893ed4c07023506ac4e834c86bcf8a343f35a4280f3480"
20
+ },
21
+ {
22
+ "path": "model.pdiparams",
23
+ "bytes": 5456137,
24
+ "sha256": "8ae7262a63a27d7011bd9ac7993bd261ba1c5dbdfae1b2753dca9711b0b4b665"
25
+ },
26
+ {
27
+ "path": "model.pdmodel",
28
+ "bytes": 10928441,
29
+ "sha256": "4a416c423b2ddb8ca0e76b8edd965051cf35981a3ac331f6a1952217e97056bb"
30
+ }
31
+ ],
32
+ "model": {
33
+ "bytes": 3150847,
34
+ "sha256": "8671c7b424d85d4f017b6410602de6095b1fbb9fffca38ad5489039dd2a0cfea"
35
+ },
36
+ "opset": 17,
37
+ "input": [
38
+ 1,
39
+ 3,
40
+ 128,
41
+ 96
42
+ ],
43
+ "output": [
44
+ 1,
45
+ 17,
46
+ 32,
47
+ 24
48
+ ],
49
+ "parameterCount": null,
50
+ "tools": {
51
+ "python": "3.11.15 (main, Jul 18 2026, 17:06:27) [MSC v.1944 64 bit (AMD64)]",
52
+ "os": "Windows-10-10.0.26200-SP0",
53
+ "paddle": "2.6.2",
54
+ "paddle2onnx": "1.3.1",
55
+ "onnx": "1.16.2",
56
+ "onnxruntime": "1.20.1",
57
+ "onnxconverterCommon": "1.16.0",
58
+ "numpy": "1.26.4",
59
+ "opencv": "4.11.0"
60
+ },
61
+ "reproduction": "tools/variants/prepare.py",
62
+ "evidence": "reports/2026-09-17-variants",
63
+ "graphModification": {
64
+ "strategy": "Conv 常量权重 float16 存储,算子前 Cast 恢复 float32",
65
+ "storagePrecision": "float16",
66
+ "activationPrecision": "float32",
67
+ "operatorPrecision": "float32",
68
+ "inputType": "float32",
69
+ "outputType": "float32",
70
+ "label": "W16A32,不宣称 FP16 计算",
71
+ "conversion": {
72
+ "constantWeights": 270,
73
+ "initializerWeights": 0,
74
+ "castNodes": 270,
75
+ "elements": 1281729
76
+ },
77
+ "graphStatistics": {
78
+ "operators": {
79
+ "Add": 79,
80
+ "ArgMax": 1,
81
+ "BatchNormalization": 268,
82
+ "Cast": 270,
83
+ "Concat": 52,
84
+ "Constant": 1511,
85
+ "Conv": 269,
86
+ "Relu": 144,
87
+ "Reshape": 99,
88
+ "Resize": 29,
89
+ "Shape": 3,
90
+ "Slice": 3,
91
+ "Split": 49,
92
+ "Transpose": 49
93
+ },
94
+ "initializerTypes": {},
95
+ "constantTensorTypes": {
96
+ "FLOAT": 1130,
97
+ "FLOAT16": 270,
98
+ "INT64": 111
99
+ },
100
+ "constantTensorElements": {
101
+ "FLOAT": 73896,
102
+ "FLOAT16": 1281729,
103
+ "INT64": 459
104
+ },
105
+ "nodes": 2826,
106
+ "initializers": 0
107
+ },
108
+ "probeOutputShape": [
109
+ 1,
110
+ 17,
111
+ 32,
112
+ 24
113
+ ],
114
+ "probeOutputType": "float32"
115
+ }
116
+ }
tinypose-128x96/0.2.0/w16a32/tinypose-128x96-w16a32.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8671c7b424d85d4f017b6410602de6095b1fbb9fffca38ad5489039dd2a0cfea
3
+ size 3150847