Research map / Detection and segmentation / 3D detection and driving
OPD: Single-View 3D Openable Part Detection
Hanxiao Jiang, Yongsen Mao, Manolis Savva, Angel X. Chang
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 object-centric (canonical coordinate) motion prediction instead of camera-centric | OpdRcnn · OPDSynth test (RGBD) | motion-averaged mAP +MAO: 53.9 -> 62.9 | Improved |
| use object-centric instead of camera-centric prediction | OpdRcnn · OPDReal test (RGBD) | motion-averaged mAP +MAO: 31.5 -> 48.1 | Improved |
| add depth to RGB input | OpdRcnn-O · OPDSynth test | part-averaged mAP +MAO: 46.9 -> 49.0 | Improved |
| use depth only instead of RGB | OpdRcnn-O · OPDSynth test | motion-averaged mAP +MAO: 61.5 -> 60.0 | Hurt |
| provide ground truth object pose (ablation) | OpdRcnn-O · OPDSynth val (RGBD) | part-averaged mAP +MAO: 47.1 -> 59.4 | Improved |
| provide ground truth part label and 2D box (ablation) | OpdRcnn-O · OPDSynth val (RGBD) | part-averaged mAP +MAO: 47.1 -> 45.4 | No clear effect |
| add depth to RGB input for door parts (ablation) | OpdRcnn-O · OPDSynth val door | part-averaged mAP +MAO: 57.0 -> 59.8 | Improved |
| add depth to RGB input for drawer parts (ablation) | OpdRcnn-O · OPDSynth val drawer | part-averaged mAP +MAO: 71.5 -> 65.9 | Hurt |
| add depth to RGB input for lid parts (ablation) | OpdRcnn-O · OPDSynth val lid | part-averaged mAP +MAO: 17.8 -> 19.7 | Improved |
| finetune synthetic-trained model on real data | OpdRcnn-O · OPDReal vs OPDSynth (RGBD) | motion-averaged mAP +MAO: 62.9 -> 48.1 | Hurt |
Closest papers by meaning
- Towards 3D Objectness Learning in an Open World (NeurIPS 2025 · similarity 0.87)
- OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detection (NeurIPS 2025 · similarity 0.86)
- OpenBox: Annotate Any Bounding Boxes in 3D (NeurIPS 2025 · similarity 0.86)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.85)
- ImOV3D: Learning Open Vocabulary Point Clouds 3D Object Detection from Only 2D Images (NeurIPS 2024 · similarity 0.85)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 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.
