Research map / 3D vision / Human pose estimation
3D Pose Estimation and 3D Model Retrieval for Objects in the Wild
Alexander Grabner, Peter M. Roth, Vincent Lepetit
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 ResNet instead of VGG as the pose-estimation backbone (ablation) | pose estimation · Pascal3D+ | MedErr: 11.7 -> 10.9 | Improved |
| use ResNet instead of VGG as the pose-estimation backbone (ablation) | pose estimation · Pascal3D+ | Acc_{π/6}: 0.8076 -> 0.8341 | Improved |
| add blur augmentation to ResNet training (ablation) | pose estimation · Pascal3D+ | Acc_{π/6}: 0.8341 -> 0.8392 | Improved |
| add blur augmentation to ResNet training (ablation) | pose estimation · Pascal3D+ | MedErr: 10.9 -> 10.9 | No clear effect |
| add blur augmentation to VGG training (ablation) | pose estimation · Pascal3D+ | Acc_{π/6}: 0.8076 -> 0.8033 | Hurt |
| add blur augmentation to VGG training (ablation) | pose estimation · Pascal3D+ | MedErr: 11.7 -> 11.6 | Improved |
| share one output layer across categories instead of using category-specific predictions (ablation) | pose estimation · Pascal3D+ | MedErr: 10.9 -> 11.5 | Hurt |
| share one output layer across categories instead of using category-specific predictions (ablation) | pose estimation · Pascal3D+ | Acc_{π/6}: 0.8392 -> 0.8133 | Hurt |
| use the predicted pose rather than a random pose for rendering (ablation) | 3D model retrieval · Pascal3D+ validation | Top-1-Acc: 0.2250 -> 0.4600 | Improved |
| use the predicted pose rather than a canonical pose for rendering (ablation) | 3D model retrieval · Pascal3D+ validation | Top-1-Acc: 0.2925 -> 0.4600 | Improved |
| use the ground-truth pose rather than the predicted pose for rendering (ablation) | 3D model retrieval · Pascal3D+ validation | Top-1-Acc: 0.4600 -> 0.4967 | Improved |
| use offline pre-computed descriptors and the closest discretized pose rather than online rendering under the predicted pose (ablation) | 3D model retrieval · Pascal3D+ validation | Top-1-Acc: 0.4600 -> 0.4375 | Hurt |
Closest papers by meaning
- In the Wild Human Pose Estimation Using Explicit 2D Features and Intermediate 3D Representations (CVPR 2019 · similarity 0.88)
- 3D Human Pose Estimation in the Wild by Adversarial Learning (CVPR 2018 · similarity 0.88)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.88)
- A Real World Dataset for Multi-View 3D Reconstruction (ECCV 2022 · similarity 0.88)
- RGBD Objects in the Wild: Scaling Real-World 3D Object Learning from RGB-D Videos (CVPR 2024 · similarity 0.87)
- Unsupervised Learning of Category-Level 3D Pose from Object-Centric Videos (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.
