Research map / Label-efficient and robust learning / Recognition and long-tail
Interspecies Knowledge Transfer for Facial Keypoint Detection
Maheen Rashid, Xiuye Gu, Yong Jae Lee
Check 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 |
|---|---|---|---|
| Add a directly supervised animal-to-human face warping network before finetuning (ablation) | Human facial keypoint detector finetuned for animal keypoint detection · Horse and sheep | failure rate | Improved |
| Add a weakly guided TPS warping network trained only with keypoint detection loss (ablation) | Human facial keypoint detector finetuned for animal keypoint detection · Horse and sheep | failure rate | Improved |
| Pretrain on human facial keypoints instead of training from scratch (ablation) | Animal facial keypoint detector · Horse and sheep | failure rate | Improved |
| Use ground-truth animal-to-human warps rather than the learned warp (ablation) | Horse keypoint detector · Horse Facial Keypoint dataset | failure rate: 8.36% -> 7.76% | Improved |
| Use the proposed method instead of TIF | Animal facial keypoint detection · Horse and sheep | failure rate: 10.44% and 12.52% points lower failure rate | Improved |
| Use the proposed method instead of simple finetuning with 500 training images | Human facial keypoint detector finetuned for horses · Horse Facial Keypoint dataset, 500 training images | failure rate: 13.39% point lower failure rate | Improved |
| Use the proposed method instead of the TPS baseline with 500 training images | Human facial keypoint detector finetuned for horses · Horse Facial Keypoint dataset, 500 training images | failure rate: 6.72% point lower failure rate | Improved |
| Increase the number of human nearest neighbors used for training (ablation) | Proposed horse keypoint detector · Horse Facial Keypoint dataset | failure rate | Mixed |
Closest papers by meaning
- Deep Cross-species Feature Learning for Animal Face Recognition via Residual Interspecies Equivariant Network (ECCV 2020 · similarity 0.87)
- Dense Interspecies Face Embedding (NeurIPS 2022 · similarity 0.87)
- AnimalWeb: A Large-Scale Hierarchical Dataset of Annotated Animal Faces (CVPR 2020 · similarity 0.87)
- Detect Any Keypoints: An Efficient Light-Weight Few-Shot Keypoint Detector (AAAI 2024 · similarity 0.83)
- SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild (NeurIPS 2026 · similarity 0.82)
- VAREN: Very Accurate and Realistic Equine Network (CVPR 2024 · 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.
