Animal Pose Estimation
Animal pose estimation is the task of identifying the pose of an animal.
( Image credit: Using DeepLabCut for 3D markerless pose estimation across species and behaviors )
Papers
Showing 1–10 of 48 papers
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | ViTPose+-H | AP | 82.4 | — | Unverified |
| 2 | ViTPose+-L | AP | 80.4 | — | Unverified |
| 3 | SuperAnimal-HRNetw32 | AP | 80.11 | — | Unverified |
| 4 | UniPose | AP | 79.2 | — | Unverified |
| 5 | ViTPose+-B | AP | 74.5 | — | Unverified |
| 6 | HRNet-w48 | AP | 73.1 | — | Unverified |
| 7 | HRNet-w32 | AP | 72.2 | — | Unverified |
| 8 | ViTPose+-S ViT-S | AP | 71.4 | — | Unverified |
| 9 | SimpleBaseline-ResNet50 | AP | 68.1 | — | Unverified |
| 10 | zero-shot SuperAnimal-HRNetw32 | AP | 68.04 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | DeepLabCut-EfficientNet-B6 | [email protected] (OOD) | 88.4 | — | Unverified |
| 2 | DeepLabCut-EfficientNet-B4 | [email protected] (OOD) | 86.9 | — | Unverified |
| 3 | DeepLabCut-RESNET-101 | [email protected] (OOD) | 84.3 | — | Unverified |
| 4 | DeepLabCut-RESNET 50 | [email protected] (OOD) | 81.3 | — | Unverified |
| 5 | DeepLabCut-MOBILENETV2-1 | [email protected] (OOD) | 77.6 | — | Unverified |
| 6 | DeepLabCut-MOBILENETV2 0.35 | [email protected] (OOD) | 63.5 | — | Unverified |
| 7 | mmpose HRNet-w32 (w/ImageNet pretrained weights) | Normalized Error (OOD) | 0.18 | — | Unverified |
| 8 | SuperAnimal-Quadruped HRNet-w32 | Normalized Error (OOD) | 0.11 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | BUCTD-CoAM-W48 (DLCRNet) | mAP | 99.1 | — | Unverified |
| 2 | SuperAnimal HRNetw32 | mAP | 98.55 | — | Unverified |
| 3 | DLCRNet | mAP | 95.8 | — | Unverified |
| 4 | ResNet50_s4graph11 | mAP | 93 | — | Unverified |
| 5 | DLCRNet_ms4graph11 | mAP | 92 | — | Unverified |
| 6 | CID-W32 | mAP | 86.8 | — | Unverified |
| 7 | zero-shot SuperAnimal HRNetw32 | mAP | 76.14 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | HRNet-W48 + Faster R-CNN | mAP | 89.1 | — | Unverified |
| 2 | BUCTD-preNet-W48 (DLCRNet) | mAP | 88.7 | — | Unverified |
| 3 | BUCTD-preNet-W48 (CID-W32) | mAP | 88 | — | Unverified |
| 4 | DLCRNet_ms4graph9 | mAP | 71.6 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | BUCTD-preNet-W48 (CID-W32) | mAP | 93.3 | — | Unverified |
| 2 | CID-W32 | mAP | 92.5 | — | Unverified |
| 3 | BUCTD-CoAM-W48 (DLCRNet) | mAP | 91.6 | — | Unverified |
| 4 | DLCRNet_ms-graph34 | mAP | 89 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | 8 Stacked Hourglass Network | [email protected] | 78.65 | — | Unverified |
| 2 | 2 Stacked Hourglass Network | [email protected] | 77.19 | — | Unverified |
| 3 | Mask R-CNN | [email protected] | 50.77 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | PARE | PA-MPJPE | 134.8 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | SuperAnimal-AnimalTokenPose | AP | 86 | — | Unverified |