Research map / Detection and segmentation / Open-set and few-shot detection
3D-DRES: Detailed 3D Referring Expression Segmentation
Qi Chen, Changli Wu, Jiayi Ji, Yiwei Ma, Liujuan Cao
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 per-layer (multi-layer) supervision of decoder outputs (ablation) | DetailBase · DetailRefer val | phrase mIoU: 50.8 -> 55.4 | Improved |
| add auxiliary score loss (ablation) | DetailBase · DetailRefer val | phrase mIoU: 55.0 -> 55.4 | Improved |
| increase decoder layers from 1 to 3 (ablation) | DetailBase · DetailRefer val | phrase mIoU: 53.1 -> 55.1 | Improved |
| increase decoder layers from 6 to 9 (ablation) | DetailBase · DetailRefer val | phrase mIoU: 55.4 -> 55.5 | No clear effect |
| joint training on ScanRefer and DetailRefer (ablation) | DetailBase · ScanRefer val (3D-RES) | mIoU: 44.0 -> 46.8 | Improved |
| joint training on ScanRefer and DetailRefer (ablation) | 3D-STMN · ScanRefer val (3D-RES) | mIoU: 41.8 -> 45.0 | Improved |
| joint training on ScanRefer and DetailRefer (ablation) | DetailBase · DetailRefer val (3D-DRES) | phrase mIoU: 55.4 -> 56.8 | Improved |
| joint training on ScanRefer and DetailRefer (ablation) | 3D-STMN · DetailRefer val (3D-DRES) | phrase mIoU: 52.7 -> 53.3 | Improved |
| one-stage query-based decoder instead of two-stage proposal matching | PNG-style two-stage adaptation · DetailRefer test | phrase mIoU: 40.4 -> 55.7 | Improved |
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.
- 3D-STMN — 3D-STMN: Dependency-Driven Superpoint-Text Matching Network for End-to-End 3D Referring Expression Segmentation (AAAI 2024) · better on 6 of 6 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
- LESS: Label-Efficient and Single-Stage Referring 3D Segmentation (NeurIPS 2024 · similarity 0.90)
- Unveiling Parts Beyond Objects: Towards Finer-Granularity Referring Expression Segmentation (CVPR 2024 · similarity 0.89)
- 3D-STMN: Dependency-Driven Superpoint-Text Matching Network for End-to-End 3D Referring Expression Segmentation (AAAI 2024 · similarity 0.89)
- AugRefer: Advancing 3D Visual Grounding via Cross-Modal Augmentation and Spatial Relation-based Referring (AAAI 2025 · similarity 0.88)
- Mask Grounding for Referring Image Segmentation (CVPR 2024 · similarity 0.88)
- ReferSplat: Referring Segmentation in 3D Gaussian Splatting (ICML 2025 · 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.
