Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- README.md +21 -1
- instanthmr.gif +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
instanthmr.gif filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -1,5 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# InstantHMR — model weights
|
| 2 |
|
|
|
|
|
|
|
| 3 |
A lightweight, ONNX-exportable distillation of
|
| 4 |
[`facebook/sam-3d-body-dinov3`](https://huggingface.co/facebook/sam-3d-body-dinov3)
|
| 5 |
for **3D human pose estimation and mesh recovery**: a RepViT-M1.5
|
|
@@ -8,7 +28,7 @@ Trained to mimic the SAM3D teacher's per-person 70-keypoint outputs from
|
|
| 8 |
a single 224×224 crop.
|
| 9 |
|
| 10 |
Demo, training, and inference code live at the main InstantHMR
|
| 11 |
-
repository.
|
| 12 |
|
| 13 |
## Files
|
| 14 |
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
license_name: sam-license
|
| 4 |
+
license_link: https://github.com/facebookresearch/sam-3d-body/blob/main/LICENSE
|
| 5 |
+
library_name: onnx
|
| 6 |
+
pipeline_tag: keypoint-detection
|
| 7 |
+
tags:
|
| 8 |
+
- human-pose-estimation
|
| 9 |
+
- human-mesh-recovery
|
| 10 |
+
- 3d-pose-estimation
|
| 11 |
+
- onnx
|
| 12 |
+
- distillation
|
| 13 |
+
- sam3d
|
| 14 |
+
- repvit
|
| 15 |
+
- cliff
|
| 16 |
+
base_model: facebook/sam-3d-body-dinov3
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
# InstantHMR — model weights
|
| 20 |
|
| 21 |
+

|
| 22 |
+
|
| 23 |
A lightweight, ONNX-exportable distillation of
|
| 24 |
[`facebook/sam-3d-body-dinov3`](https://huggingface.co/facebook/sam-3d-body-dinov3)
|
| 25 |
for **3D human pose estimation and mesh recovery**: a RepViT-M1.5
|
|
|
|
| 28 |
a single 224×224 crop.
|
| 29 |
|
| 30 |
Demo, training, and inference code live at the main InstantHMR
|
| 31 |
+
repository: <https://github.com/mohamdev/InstantHMR>.
|
| 32 |
|
| 33 |
## Files
|
| 34 |
|
instanthmr.gif
ADDED
|
Git LFS Details
|