Research map / 3D vision / Human pose estimation
Occlusion-Net: 2D/3D Occluded Keypoint Localization Using Graph Networks
N. Dinesh Reddy, Minh Vo, Srinivasa G. Narasimhan
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 |
|---|---|---|---|
| Use Occlusion-Net instead of [25] | KITTI-3D visible keypoints · KITTI-3D visible keypoints, Truncation | PCK@0.1: 78.5 -> 87.41 | Improved |
| Use Occlusion-Net instead of [25] | KITTI-3D visible keypoints · KITTI-3D visible keypoints, Oth-Occ | PCK@0.1: 85.3 -> 86.45 | Improved |
| Use Occlusion-Net instead of [25] | KITTI-3D visible keypoints · KITTI-3D visible keypoints, Car-Occ | PCK@0.1: 82.9 -> 81.68 | Hurt |
| Use Occlusion-Net instead of [25] | KITTI-3D visible keypoints · KITTI-3D visible keypoints, All | PCK@0.1: 85.0 -> 88.8 | Improved |
| Use Occlusion-Net instead of [25] | KITTI-3D visible keypoints · KITTI-3D visible keypoints, 3D Full | 3D keypoint localization: 95.3 -> 93.2 | Hurt |
| Use Occlusion-Net instead of [25] | KITTI-3D visible keypoints · KITTI-3D visible keypoints | yaw error: 2.2 -> 1.9 | Improved |
| Add a graph encoder for occlusion prediction (ablation) | MaskRCNN keypoint detector · CarFusion occlusion prediction | AUC: 0.83 -> 0.95 | Improved |
| Add graph-based keypoint localization (ablation) | MaskRCNN · Car-render, 3 invisible points | PCK: 22% | Improved |
| Add graph-based keypoint localization (ablation) | MaskRCNN · Car-render, 9 invisible points | PCK: 10% | Improved |
| Add graph-based keypoint localization (ablation) | MaskRCNN · CarFusion, 3 occluded points | PCK: 6% | Improved |
| Add graph-based keypoint localization (ablation) | MaskRCNN · CarFusion, severe occlusion | PCK: 10% or more | Improved |
| Add 3D-KGNN instead of relying only on 2D-KGNN (ablation) | Occlusion-Net · Car-render, varying number of invisible points | PCK: variance of 5%; 15%; 25% | Mixed |
Closest papers by meaning
- Few-Shot Geometry-Aware Keypoint Localization (CVPR 2023 · similarity 0.90)
- Deep Supervision With Shape Concepts for Occlusion-Aware 3D Object Parsing (CVPR 2017 · similarity 0.88)
- Key-Grid: Unsupervised 3D Keypoints Detection using Grid Heatmap Features (NeurIPS 2024 · similarity 0.87)
- Back to 3D: Few-Shot 3D Keypoint Detection with Back-Projected 2D Features (CVPR 2024 · similarity 0.86)
- Occlusion-aware Text-Image-Point Cloud Pretraining for Open-World 3D Object Recognition (CVPR 2025 · similarity 0.86)
- DGC-GNN: Leveraging Geometry and Color Cues for Visual Descriptor-Free 2D-3D Matching (CVPR 2024 · similarity 0.85)
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.
