Research map / 3D vision / Human pose estimation
OpenMonkeyChallenge: Dataset and Benchmark Challenges for Pose Estimation of Non-human Primates
Yuan Yao, Praneet Bala, Abhiraj Mohan, Eliza Bliss‐Moreau, Kristine Coleman, Sienna M. Freeman, Christopher J. Machado, Jessica Raper, Jan Zimmermann, Benjamin Y. Hayden, Hyun Soo Park
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train on OpenMonkeyChallenge instead of retraining nothing (COCO human model) | CPM · OpenMonkeyChallenge test | MPJPE: 0.197 -> 0.074 | Improved |
| train on OpenMonkeyChallenge instead of COCO human model | CPM · OpenMonkeyChallenge test | PCK@0.2: 0.265 -> 0.761 | Improved |
| primate model vs human model on its own domain (gap) | CPM · PCK@0.2 | PCK@0.2: 0.849 -> 0.761 | Hurt |
| increase training data from 20% to 100% (ablation) | CPM · OpenMonkeyChallenge test | PCK@0.2: 15% gain at PCK@0.2 | Improved |
| train on OpenMonkeyChallenge instead of MacaquePose/OpenMonkeyPose | CPM · cross-dataset (OMC, MacaquePose, OpenMonkeyPose tests) | MPJPE | Improved |
| train on OpenMonkeyPose (lab data) instead of OpenMonkeyChallenge | CPM · OpenMonkeyChallenge and MacaquePose tests | MPJPE | Hurt |
| use top-down model instead of bottom-up | HRNet-W48 vs HigherHRNet-W48 · OpenMonkeyChallenge test | mean MPJPE: 0.092 -> 0.064 | Improved |
| increase backbone width from W32 to W48 (ablation) | HRNet · OpenMonkeyChallenge test | AP: 70.7 -> 76.5 | Improved |
| increase backbone width from W32 to W48 (ablation) | HigherHRNet · OpenMonkeyChallenge test | AP: 59.1 -> 65.3 | Improved |
| increase depth from ResNet101 to ResNet152 (ablation) | SimpleBaseline · OpenMonkeyChallenge test | mean PCK@0.2: 0.863 -> 0.901 | Improved |
| choice among top CNN architectures | pose estimators · OpenMonkeyChallenge test | mean PCK@0.2: 0.890 to 0.903 | No clear effect |
Closest papers by meaning
- BigMaQ: A Big Macaque Motion and Animation Dataset Bridging Image and 3D Pose Representations (ICLR 2026 · similarity 0.87)
- PanAf20K: A Large Video Dataset for Wild Ape Detection and Behaviour Recognition (IJCV 2024 · similarity 0.85)
- ChimpACT: A Longitudinal Dataset for Understanding Chimpanzee Behaviors (NeurIPS 2023 · similarity 0.85)
- Learning Structure-Supporting Dependencies via Keypoint Interactive Transformer for General Mammal Pose Estimation (IJCV 2025 · similarity 0.85)
- Sapiens: Foundation for Human Vision Models (ECCV 2024 · similarity 0.85)
- SapiensID: Foundation for Human Recognition (CVPR 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
