Research map / 3D vision / Human pose estimation
StarMap for Category-Agnostic Keypoint and Viewpoint Estimation
Xingyi Zhou, Arjun Karpur, Linjie Luo, Qixing Huang
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 |
|---|---|---|---|
| use ground-truth StarMap instead of predicted (ablation) | StarMap viewpoint pipeline · Pascal3D+ viewpoint | mean Acc@pi/6: 0.8225 -> 0.8211 | No clear effect |
| use ground-truth CanViewFeature instead of predicted (ablation) | StarMap viewpoint pipeline · Pascal3D+ viewpoint | mean Acc@pi/6: 0.8225 -> 0.8651 | Improved |
| use ground-truth DepthMap instead of predicted (ablation) | StarMap viewpoint pipeline · Pascal3D+ viewpoint | mean Median Error: 10.43 -> 4.7 | Improved |
| use ground-truth DepthMap instead of predicted (ablation) | StarMap viewpoint pipeline · Pascal3D+ viewpoint | mean Acc@pi/6: 0.8225 -> 0.8637 | Improved |
| use DepthMap lifting rather than direct PnP alignment | StarMap viewpoint estimator · Pascal3D+ viewpoint | mean Acc@pi/6: 0.7416 -> 0.8225 | Improved |
| use DepthMap lifting rather than direct PnP alignment | StarMap viewpoint estimator · Pascal3D+ viewpoint | mean Acc@pi/18: 0.4643 -> 0.4818 | Improved |
| use keypoint alignment rather than category-general bin classification | viewpoint estimator · Pascal3D+ viewpoint | mean Acc@pi/18: 0.3502 -> 0.4818 | Improved |
| use category-specific classifiers rather than a general one | Res18 viewpoint classifier · Pascal3D+ viewpoint | mean Acc@pi/6: 0.7875 -> 0.8121 | No clear effect |
| use CanViewFeature instead of SIFT or Conv5 features | keypoint classification · Pascal3D+ keypoint classification | mean accuracy: 57 -> 81 | Improved |
| use single-channel StarMap instead of 16-channel Hourglass output | HourglassNetwork · MPII (oracle ID) | PCKh@0.5 total: 93.22 -> 88.17 | Hurt |
| use single-channel StarMap instead of 16-channel Hourglass output | HourglassNetwork · MPII (learned ID) | PCKh@0.5 total: 88.39 -> 79.85 | Hurt |
| replace depth sub-network with DepthMap channels | Zhou et al. 3D pose · Human3.6M | MPJPE (mm): 64.9 -> 65.77 | No clear effect |
Closest papers by meaning
- KeypointDETR: An End-to-End 3D Keypoint Detector (ECCV 2024 · similarity 0.89)
- Unsupervised 3D Structure Inference from Category-Specific Image Collections (CVPR 2024 · similarity 0.88)
- Key-Grid: Unsupervised 3D Keypoints Detection using Grid Heatmap Features (NeurIPS 2024 · similarity 0.88)
- Back to 3D: Few-Shot 3D Keypoint Detection with Back-Projected 2D Features (CVPR 2024 · similarity 0.88)
- ESCAPE: Encoding Super-keypoints for Category-Agnostic Pose Estimation (CVPR 2024 · similarity 0.86)
- Meta-Point Learning and Refining for Category-Agnostic Pose Estimation (CVPR 2024 · similarity 0.86)
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.
