Papers › SuperAnimal pretrained pose estimation models for behavioral analysis
SuperAnimal pretrained pose estimation models for behavioral analysis
Shaokai Ye, Anastasiia Filippova, Jessy Lauer, Steffen Schneider, Maxime Vidal, Tian Qiu, Alexander Mathis, Mackenzie Weygandt Mathis
Quantification of behavior is critical in applications ranging from neuroscience, veterinary medicine and animal conservation efforts. A common key step for behavioral analysis is first extracting relevant keypoints on animals, known as pose estimation. However, reliable inference of poses currently requires domain knowledge and manual labeling effort to build supervised models. We present a series of technical innovations that enable a new method, collectively called SuperAnimal, to develop unified foundation models that can be used on over 45 species, without additional human labels. Concretely, we introduce a method to unify the keypoint space across differently labeled datasets (via our generalized data converter) and for training these diverse datasets in a manner such that they don't catastrophically forget keypoints given the unbalanced inputs (via our keypoint gradient masking and memory replay approaches). These models show excellent performance across six pose benchmarks. Then, to ensure maximal usability for end-users, we demonstrate how to fine-tune the models on differently labeled data and provide tooling for unsupervised video adaptation to boost performance and decrease jitter across frames. If the models are fine-tuned, we show SuperAnimal models are 10-100× more data efficient than prior transfer-learning-based approaches. We illustrate the utility of our models in behavioral classification in mice and gait analysis in horses. Collectively, this presents a data-efficient solution for animal pose estimation.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Datasets
Introduced by this paper, per the archive.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| 2D Pose Estimation | iRodent | fine-tuned HRNetw32 pretrained on SuperAnimal (1 fac of data) | Average mAP | 72.971 | #1 of 8 | Archive leaderboard | report |
| 2D Pose Estimation | iRodent | fine-tuned HRNetw32 pretrained on AP-10K (1 fac of data) | Average mAP | 61.635 | #2 of 8 | Archive leaderboard | report |
| 2D Pose Estimation | iRodent | fine-tuned HRNetw32 pretrained on SuperAnimal (0.01 fac of data) | Average mAP | 60.853 | #3 of 8 | Archive leaderboard | report |
| 2D Pose Estimation | iRodent | fine-tuned HRNetw32 pretrained on ImageNet | Average mAP | 58.857 | #4 of 8 | Archive leaderboard | report |
| 2D Pose Estimation | iRodent | zero-shot HRNet-w32 pretrained on SuperAnimal-Quadruped | Average mAP | 58.557 | #5 of 8 | Archive leaderboard | report |
| 2D Pose Estimation | iRodent | zero-shot AnimalTokenPose pretrained on AP-10K | Average mAP | 55.415 | #6 of 8 | Archive leaderboard | report |
| 2D Pose Estimation | iRodent | fine-tuned HRNetw32 pretrained on AP-10K (0.01 fac of data) | Average mAP | 43.144 | #7 of 8 | Archive leaderboard | report |
| 2D Pose Estimation | iRodent | zero-shot HRNet-w32 pretrained on AP-10K | Average mAP | 40.389 | #8 of 8 | Archive leaderboard | report |
| Animal Pose Estimation | AP-10K | SuperAnimal-HRNetw32 | AP | 80.113 | #3 of 10 | Archive leaderboard | report |
| Animal Pose Estimation | AP-10K | zero-shot SuperAnimal-HRNetw32 | AP | 68.038 | #10 of 10 | Archive leaderboard | report |
| Animal Pose Estimation | Animal-Pose Dataset | SuperAnimal-AnimalTokenPose | AP | 86 | #1 of 1 | Archive leaderboard | report |
| Animal Pose Estimation | Horse-10 | SuperAnimal-Quadruped HRNet-w32 | Normalized Error (OOD) | 0.1091 | #7 of 8 | Archive leaderboard | report |
| Animal Pose Estimation | Horse-10 | mmpose HRNet-w32 (w/ImageNet pretrained weights) | Normalized Error (OOD) | 0.179 | #8 of 8 | Archive leaderboard | report |
| Animal Pose Estimation | TriMouse-161 | SuperAnimal HRNetw32 | mAP | 98.547 | #2 of 7 | Archive leaderboard | report |
| Animal Pose Estimation | TriMouse-161 | zero-shot SuperAnimal HRNetw32 | mAP | 76.139 | #7 of 7 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections