Research map / Detection and segmentation / Weakly supervised detection and segmentation
DQS3D: Densely-matched Quantization-aware Semi-supervised 3D Detection
Huan-ang Gao, Beiwen Tian, Pengfei Li, Hao Zhao, Guyue Zhou
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 |
|---|---|---|---|
| add quantization error correction (QEC) (ablation) | DQS3D · ScanNet v2 (20% labels), voxel 0.01m | mAP@0.50: 47.1 -> 48.5 | Improved |
| add QEC at larger voxel size 0.03m (ablation) | DQS3D · ScanNet v2 (20% labels), voxel 0.03m | weighted IoU (%): 65.63 -> 68.12 | Improved |
| add box consistency loss (ablation) | DQS3D · ScanNet v2 (20% labels) | mAP@0.50: 45.3 -> 48.5 | Improved |
| add centerness consistency loss (ablation) | DQS3D · ScanNet v2 (20% labels) | mAP@0.50: 46.4 -> 48.5 | Improved |
| add class consistency loss (ablation) | DQS3D · ScanNet v2 (20% labels) | mAP@0.50: 47.3 -> 48.5 | Improved |
| use dense matching instead of sparse proposal matching (ablation) | FCAF3D mean teacher · ScanNet v2 (20% labels) | mAP@0.50: 44.2 -> 48.5 | Improved |
| apply dense matching to a different backbone (TR3D) (ablation) | TR3D · ScanNet v2 (20% labels) | mAP@0.50: 46.8 -> 49.9 | Improved |
| apply DQS3D self-training in the fully supervised setting | FCAF3D · ScanNet v2 (100% labels) | mAP@0.50: 55.1 -> 56.3 | Improved |
| apply DQS3D self-training in the fully supervised setting | FCAF3D · SUN RGB-D (100% labels) | mAP@0.50: 47.5 -> 48.2 | Improved |
| DQS3D vs 3DIoUMatch with 20% labels | semi-supervised 3D detection · ScanNet v2 (20% labels) | mAP@0.50: 35.2 -> 48.5 | Improved |
Later papers that compared with this method
1 later accepted paper reports DQS3D as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Learning Class Prototypes for Unified Sparse-Supervised 3D Object Detection (CVPR 2025 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- 3DIoUMatch — 3DIoUMatch: Leveraging IoU Prediction for Semi-Supervised 3D Object Detection (CVPR 2021) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Leveraging Temporal Cues for Semi-Supervised Multi-View 3D Object Detection (CVPR 2025 · similarity 0.90)
- Semi-supervised 3D Object Detection with PatchTeacher and PillarMix (AAAI 2024 · similarity 0.89)
- Dual-Perspective Knowledge Enrichment for Semi-supervised 3D Object Detection (AAAI 2024 · similarity 0.89)
- Decoupled Pseudo-labeling for Semi-Supervised Monocular 3D Object Detection (CVPR 2024 · similarity 0.87)
- Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 3D Object Detection (AAAI 2024 · similarity 0.87)
- A-Teacher: Asymmetric Network for 3D Semi-Supervised Object Detection (CVPR 2024 · similarity 0.87)
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.
