Research map / 3D vision / Face and shape
Shape and Viewpoint without Keypoints
Shubham Goel, Angjoo Kanazawa, Jitendra Malik
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 |
|---|---|---|---|
| add camera-multiplex (K camera hypotheses per image) (ablation) | CMR without keypoints (CMR-nokp) · CUB-200-2011 test | rotation error (deg): 87.52 -> 45.52 | Improved |
| add camera-multiplex (ablation) | CMR-nokp · CUB-200-2011 test | azimuth-elevation entropy (nats): 5.73 -> 7.26 | Improved |
| add camera-multiplex (ablation) | CMR-nokp · CUB-200-2011 test | Wasserstein distance azimuth (deg): 64.66 -> 8.66 | Improved |
| add texture prediction loss (ablation) | U-CMR (noTex) · CUB-200-2011 test | rotation error (deg): 61.82 -> 45.52 | Improved |
| add texture prediction loss (ablation) | U-CMR (noTex) · CUB-200-2011 test | azimuth-elevation entropy (nats): 7.36 -> 7.26 | No clear effect |
| add texture prediction loss (ablation) | U-CMR (noTex) · CUB-200-2011 test | Wasserstein distance azimuth (deg): 16.08 -> 8.66 | Improved |
| add texture prediction loss (ablation) | U-CMR (noTex) · CUB-200-2011 test | Wasserstein distance elevation (deg): 7.90 -> 6.50 | Improved |
| use direct texture generation via decoder instead of texture-flow (ablation) | U-CMR · CUB-200-2011 | camera pose collapse | Improved |
| use keypoint-free U-CMR instead of keypoint-supervised CMR | 3D reconstruction · PASCAL 3D+ car test | mean IoU: 0.640 -> 0.646 | No clear effect |
| use keypoint-free U-CMR instead of CMR | camera estimation · CUB-200-2011 test | rotation error (deg): 22.94 -> 45.52 | Hurt |
Closest papers by meaning
- Learning Category-Specific Mesh Reconstruction from Image Collections (ECCV 2018 · similarity 0.92)
- Shelf-Supervised Mesh Prediction in the Wild (CVPR 2021 · similarity 0.89)
- Self-supervised Single-view 3D Reconstruction via Semantic Consistency (ECCV 2020 · similarity 0.89)
- Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading (IJCV 2019 · similarity 0.89)
- Few-Shot Generalization for Single-Image 3D Reconstruction via Priors (ICCV 2019 · similarity 0.89)
- SAOR: Single-View Articulated Object Reconstruction (CVPR 2024 · similarity 0.88)
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.
