Research map / Detection and segmentation / 3D detection and driving
3DIoUMatch: Leveraging IoU Prediction for Semi-Supervised 3D Object Detection
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Apply 3DIoUMatch pseudo-label filtering and IoU-aware localization handling instead of SESS | VoteNet · ScanNet val, 10% labeled | mAP@0.25: 39.5±1.8 -> 47.2±0.4; +7.7 | Helped |
| Apply 3DIoUMatch pseudo-label filtering and IoU-aware localization handling instead of SESS | VoteNet · ScanNet val, 10% labeled | mAP@0.5: 19.8±1.3 -> 28.3±1.5; +8.5 | Helped |
| Apply 3DIoUMatch pseudo-label filtering and IoU-aware localization handling instead of SESS | VoteNet · SUN-RGBD val, 5% labeled | mAP@0.25: 34.2±2.0 -> 39.0±1.9; +4.8 | Helped |
| Apply 3DIoUMatch pseudo-label filtering and IoU-aware localization handling instead of SESS | VoteNet · SUN-RGBD val, 5% labeled | mAP@0.5: 13.1±1.0 -> 21.1±1.7; +8.0 | Helped |
| Add objectness and class confidence filtering to pseudo-labels (ablation) | VoteNet · ScanNet 10% labeled | mAP@0.25: 38.4 -> 44.5 | Helped |
| Add train-time objectness-based NMS after objectness/class filtering (ablation) | VoteNet · ScanNet 10% labeled | mAP@0.25: 44.5 -> 44.2 | Hurt |
| Add IoU filtering and IoU-guided NMS in place of objectness-only filtering/NMS (ablation) | VoteNet · ScanNet 10% labeled | mAP@0.25: 44.2 -> 45.9 | Helped |
| Use IoU-guided Lower-Half Suppression instead of IoU-guided NMS for train-time deduplication (ablation) | VoteNet · ScanNet 10% labeled | mAP@0.25: 45.9 -> 46.5 | Helped |
| Use IoU-guided Lower-Half Suppression instead of IoU-guided NMS for train-time deduplication (ablation) | VoteNet · SUN-RGBD 5% labeled | mAP@0.25: 37.4 -> 37.9 | Helped |
| Add test-time IoU-guided NMS (ablation) | VoteNet · ScanNet 10% labeled | mAP@0.5: 26.9 -> 28.2 | Helped |
| Add test-time IoU optimization after IoU-guided NMS (ablation) | VoteNet · ScanNet 10% labeled | mAP@0.25: 47.0 -> 47.2 | Helped |
| Add vote supervision on unlabeled data using pseudo-labels (ablation) | 3DIoUMatch VoteNet · ScanNet 10% labeled | mAP@0.25: 47.2 -> 45.4 | Hurt |
Closest papers by meaning
- Leveraging Temporal Cues for Semi-Supervised Multi-View 3D Object Detection (CVPR 2025 · similarity 0.90)
- Dual-Perspective Knowledge Enrichment for Semi-supervised 3D Object Detection (AAAI 2024 · similarity 0.90)
- Semi-supervised 3D Object Detection with PatchTeacher and PillarMix (AAAI 2024 · similarity 0.89)
- ImOV3D: Learning Open Vocabulary Point Clouds 3D Object Detection from Only 2D Images (NeurIPS 2024 · similarity 0.88)
- Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 3D Object Detection (AAAI 2024 · similarity 0.88)
- UNION: Unsupervised 3D Object Detection using Object Appearance-based Pseudo-Classes (NeurIPS 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.