card: update from fastforks 24802779
Browse files
README.md
CHANGED
|
@@ -18,7 +18,7 @@ Selective-IGEV (Wang et al. 2024, Selective-Stereo): IGEV-Stereo with selective
|
|
| 18 |
|
| 19 |
This repo is the model card only: it holds no weights. fastforks downloads the authors' weights from the links below and checks them against the SHA-256 in the model card. Code: <https://github.com/kleinicke/fastforks>. Kernels for the `kernels` backend: <https://huggingface.co/fastforks/kernels>. Original code: <https://github.com/Windsrain/Selective-Stereo> at `c39acbfef9fb`.
|
| 20 |
|
| 21 |
-
Task: stereo. Tier:
|
| 22 |
|
| 23 |
Input: a rectified left / right image pair. Output: a disparity map (pixels, left view).
|
| 24 |
|
|
@@ -67,9 +67,15 @@ Tested up to (portable, fp16, RTX 5060 Ti 16 GB): `selective-igev-eth3d-finetune
|
|
| 67 |
|
| 68 |
### torch.compile and TensorRT
|
| 69 |
|
| 70 |
-
| model | input | portable
|
| 71 |
-
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
| 72 |
-
| `selective-igev-eth3d-finetune` | 994x1482 | 998 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
|
| 74 |
### Fixed-size speed and peak memory, all models
|
| 75 |
|
|
@@ -90,23 +96,17 @@ Tested up to (portable, fp16, RTX 5060 Ti 16 GB): `selective-igev-eth3d-finetune
|
|
| 90 |
| `selective-igev-sceneflow` | portable | 190.6 ms / 282 / 362 | 575.4 ms / 769 / 1036 | 1420.3 ms / 1690 / 2030 | 8917.6 ms (n=3) / 10549 / 14320 |
|
| 91 |
| `selective-igev-sceneflow` | kernels | 82.5 ms / 260 / 368 | 213.5 ms / 693 / 1010 | 473.3 ms / 1502 / 2046 | 2808.4 ms (n=3) / 6928 / 8800 |
|
| 92 |
|
| 93 |
-
### Apple silicon (MPS)
|
| 94 |
-
|
| 95 |
-
| model | `portable` | `kernels` | speed-up |
|
| 96 |
-
| --- | --- | --- | --- |
|
| 97 |
-
| `selective-igev-sceneflow` | 2292 ms / 1727 MiB (float16) | 1878 ms / 1704 MiB | 1.22x |
|
| 98 |
-
|
| 99 |
Fixed-size cells: time (ms) / peak allocated / peak reserved memory (MiB).
|
| 100 |
|
| 101 |
- `selective-igev-eth3d-train`, `selective-igev-kitti12`, `selective-igev-kitti15`, `selective-igev-middlebury-finetune`, `selective-igev-middlebury-train`, `selective-igev-sceneflow`: same network and settings as `selective-igev-eth3d-finetune`, measured once (rows not repeated).
|
| 102 |
|
| 103 |
-
Hardware, protocol and caveats: [benchmarks](https://github.com/kleinicke/fastforks/blob/main/docs/benchmarks.md) (RTX 5060 Ti 16 GB, sm_120). Original = the
|
| 104 |
|
| 105 |
## Accuracy
|
| 106 |
|
| 107 |
### On public datasets
|
| 108 |
|
| 109 |
-
Original = the
|
| 110 |
|
| 111 |
| variant | dataset | original fp32 | original fp16 | portable | kernels | note |
|
| 112 |
|---|---|---|---|---|---|---|
|
|
@@ -125,7 +125,7 @@ Original = the authors' code, same checkpoint and protocol.
|
|
| 125 |
| `selective-igev-eth3d-train` | ETH3D two-view train, bad > 1 px % / EPE | 0.78 / 0.137 | 0.78 / 0.137 | 0.78 / 0.137 | 0.78 / 0.137 | trained on this dataset's public images; not a zero-shot result |
|
| 126 |
| `selective-igev-eth3d-train` | Middlebury 2014 trainingH, bad > 2 px % / EPE | 7.96 / 2.158 | 7.97 / 2.153 | 7.99 / 2.161 | 8.01 / 2.158 | |
|
| 127 |
|
| 128 |
-
Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-05.
|
| 129 |
|
| 130 |
### Parity with the original code
|
| 131 |
|
|
@@ -152,4 +152,4 @@ Install with `pip install fastforks` (see <https://github.com/kleinicke/fastfork
|
|
| 152 |
|
| 153 |
`stereo.load("selective-igev-eth3d-finetune", mode="kernels")` runs our CUDA kernels (`igev`, `cutlass_conv`, `cudnn_hacc`). It needs a prebuilt kernel for your GPU architecture (downloaded from <https://huggingface.co/fastforks/kernels> on first use) and a recent NVIDIA driver; if there is none it raises `BackendUnavailableError` with the reason and never falls back silently. `mode="auto"` picks `kernels` when it can run and logs the choice.
|
| 154 |
|
| 155 |
-
<!-- staged from fastforks
|
|
|
|
| 18 |
|
| 19 |
This repo is the model card only: it holds no weights. fastforks downloads the authors' weights from the links below and checks them against the SHA-256 in the model card. Code: <https://github.com/kleinicke/fastforks>. Kernels for the `kernels` backend: <https://huggingface.co/fastforks/kernels>. Original code: <https://github.com/Windsrain/Selective-Stereo> at `c39acbfef9fb`.
|
| 20 |
|
| 21 |
+
Task: stereo. Tier: stable. Outputs: `disparity`. Backends: `portable`, `kernels`.
|
| 22 |
|
| 23 |
Input: a rectified left / right image pair. Output: a disparity map (pixels, left view).
|
| 24 |
|
|
|
|
| 67 |
|
| 68 |
### torch.compile and TensorRT
|
| 69 |
|
| 70 |
+
| model | input | time: portable | compile | TensorRT fp16 | kernels | preparation: compile (cold first call) | ONNX + engine build | peak MiB (alloc / reserved): portable | compile | kernels | TensorRT (engine + torch) |
|
| 71 |
+
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
| 72 |
+
| `selective-igev-eth3d-finetune` | 994x1482 | 998 ms | 691 ms | 461 ms | 349 ms | 126 s | 256 s | 1238 / 1488 | 919 / 1088 | 1109 / 1390 | 1279 / 1316 |
|
| 73 |
+
| `selective-igev-eth3d-train` | 994x1482 | 998 ms | 691 ms | 462 ms | 349 ms | 127 s | 260 s | 1238 / 1488 | 915 / 1028 | 1109 / 1390 | 1278 / 1334 |
|
| 74 |
+
| `selective-igev-kitti12` | 994x1482 | 998 ms | 691 ms | 462 ms | 349 ms | 125 s | 255 s | 1238 / 1488 | 915 / 1028 | 1109 / 1390 | 1278 / 1334 |
|
| 75 |
+
| `selective-igev-kitti15` | 994x1482 | 998 ms | 691 ms | 462 ms | 349 ms | 127 s | 256 s | 1238 / 1488 | 915 / 1028 | 1109 / 1390 | 1278 / 1334 |
|
| 76 |
+
| `selective-igev-middlebury-finetune` | 994x1482 | 998 ms | 692 ms | 462 ms | 349 ms | 127 s | 258 s | 1238 / 1488 | 915 / 1028 | 1109 / 1390 | 1280 / 1316 |
|
| 77 |
+
| `selective-igev-middlebury-train` | 994x1482 | 998 ms | 692 ms | 462 ms | 349 ms | 128 s | 256 s | 1238 / 1488 | 915 / 1028 | 1109 / 1390 | 1278 / 1334 |
|
| 78 |
+
| `selective-igev-sceneflow` | 994x1482 | 998 ms | 692 ms | 463 ms | 349 ms | 127 s | 255 s | 1238 / 1488 | 915 / 1028 | 1109 / 1390 | 1278 / 1334 |
|
| 79 |
|
| 80 |
### Fixed-size speed and peak memory, all models
|
| 81 |
|
|
|
|
| 96 |
| `selective-igev-sceneflow` | portable | 190.6 ms / 282 / 362 | 575.4 ms / 769 / 1036 | 1420.3 ms / 1690 / 2030 | 8917.6 ms (n=3) / 10549 / 14320 |
|
| 97 |
| `selective-igev-sceneflow` | kernels | 82.5 ms / 260 / 368 | 213.5 ms / 693 / 1010 | 473.3 ms / 1502 / 2046 | 2808.4 ms (n=3) / 6928 / 8800 |
|
| 98 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
Fixed-size cells: time (ms) / peak allocated / peak reserved memory (MiB).
|
| 100 |
|
| 101 |
- `selective-igev-eth3d-train`, `selective-igev-kitti12`, `selective-igev-kitti15`, `selective-igev-middlebury-finetune`, `selective-igev-middlebury-train`, `selective-igev-sceneflow`: same network and settings as `selective-igev-eth3d-finetune`, measured once (rows not repeated).
|
| 102 |
|
| 103 |
+
Hardware, protocol and caveats: [benchmarks](https://github.com/kleinicke/fastforks/blob/main/docs/benchmarks.md) (RTX 5060 Ti 16 GB, sm_120). Original = the original code named above at the precision of portable (fp16, or `(fp32)` where fp16 is unsupported or not finite); original fp32 = its default fp32 run. Peak memory is the torch allocator's.
|
| 104 |
|
| 105 |
## Accuracy
|
| 106 |
|
| 107 |
### On public datasets
|
| 108 |
|
| 109 |
+
Original = the original code named under Details, same checkpoint and protocol.
|
| 110 |
|
| 111 |
| variant | dataset | original fp32 | original fp16 | portable | kernels | note |
|
| 112 |
|---|---|---|---|---|---|---|
|
|
|
|
| 125 |
| `selective-igev-eth3d-train` | ETH3D two-view train, bad > 1 px % / EPE | 0.78 / 0.137 | 0.78 / 0.137 | 0.78 / 0.137 | 0.78 / 0.137 | trained on this dataset's public images; not a zero-shot result |
|
| 126 |
| `selective-igev-eth3d-train` | Middlebury 2014 trainingH, bad > 2 px % / EPE | 7.96 / 2.158 | 7.97 / 2.153 | 7.99 / 2.161 | 8.01 / 2.158 | |
|
| 127 |
|
| 128 |
+
Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 468fba10, 2026-11-05.
|
| 129 |
|
| 130 |
### Parity with the original code
|
| 131 |
|
|
|
|
| 152 |
|
| 153 |
`stereo.load("selective-igev-eth3d-finetune", mode="kernels")` runs our CUDA kernels (`igev`, `cutlass_conv`, `cudnn_hacc`). It needs a prebuilt kernel for your GPU architecture (downloaded from <https://huggingface.co/fastforks/kernels> on first use) and a recent NVIDIA driver; if there is none it raises `BackendUnavailableError` with the reason and never falls back silently. `mode="auto"` picks `kernels` when it can run and logs the choice.
|
| 154 |
|
| 155 |
+
<!-- staged from fastforks 24802779 (2026-11-15) -->
|