Research map / 3D vision / 3D generation
Seeing Behind Objects for 3D Multi-Object Tracking in RGB-D Sequences
Norman Muller, Yu-Shiang Wong, Niloy J. Mitra, Angela Dai, Matthias Niessner
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 |
|---|---|---|---|
| add object completion while retaining canonical-correspondence matching (ablation) | Ours · DynSynth test sequences | MOTA (%): further improves the tracking performance by 6.7% mean MOTA | Improved |
| add object completion with bounding-box-overlap matching (ablation) | Ours (no corr.) · DynSynth test sequences | MOTA (%): 1.6% mean MOTA | Improved |
| use canonical-correspondence association instead of bounding-box-only association, without completion (ablation) | Ours (no compl.) · DynSynth test sequences | MOTA (%): 5% mean MOTA | Improved |
| add object completion (ablation) | Ours · DynSynth individual RGB-D frames | Median rotation error: 7.4$^{\circ}$ -> 5.7$^{\circ}$ | Improved |
| add object completion (ablation) | Ours · DynSynth individual RGB-D frames | Median translation error: 15.4cm -> 12.3cm | Improved |
| add object completion (ablation) | Ours · ScanNet+Scan2CAD individual RGB-D frames | Median rotation error: 16.6$^{\circ}$ -> 13.3$^{\circ}$ | Improved |
| add object completion (ablation) | Ours · ScanNet+Scan2CAD individual RGB-D frames | Median translation error: 22.0cm -> 18.3cm | Improved |
Closest papers by meaning
- Towards Generic 3D Tracking in RGBD Videos: Benchmark and Baseline (ECCV 2022 · similarity 0.91)
- Tracking by 3D Model Estimation of Unknown Objects in Videos (ICCV 2023 · similarity 0.90)
- Real-Time Tracking of Single and Multiple Objects from Depth-Colour Imagery Using 3D Signed Distance Functions (IJCV 2017 · similarity 0.89)
- KV-Tracker: Real-Time Pose Tracking with Transformers (CVPR 2026 · similarity 0.89)
- SpatialTrackerV2: Advancing 3D Point Tracking with Explicit Camera Motion (ICCV 2025 · similarity 0.89)
- Joint Spatial-Temporal Optimization for Stereo 3D Object Tracking (CVPR 2020 · 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.
