Research map / Point clouds / Point cloud classification
EDA: Explicit Text-Decoupling and Dense Alignment for 3D Visual Grounding
Yanmin Wu, Xinhua Cheng, Renrui Zhang, Zesen Cheng, Jian Zhang
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 explicit text decoupling and dense alignment instead of sparse main-object-word alignment | EDA · ScanRefer Overall | Acc@0.25IoU: 50.42 -> 54.59 | Improved |
| Use explicit text decoupling and dense alignment instead of sparse main-object-word alignment | EDA · SR3D and NR3D | Acc@0.25IoU: 65.6 -> 68.1; 51.5 -> 52.1 | Improved |
| Add position-aligned loss to semantic-aligned loss (ablation) | EDA · ScanRefer | Acc@0.25IoU: 52.2 -> 54.6 | Improved |
| Add semantic-aligned loss to position-aligned loss (ablation) | EDA · ScanRefer | Acc@0.25IoU: 51.2 -> 54.6 | Improved |
| Add Attribute, Pronoun, Relationship, and Auxiliary object components to Main object alignment (ablation) | EDA · ScanRefer | Acc@0.25IoU: 51.5 -> 54.6 | Improved |
| Add the Relationship component to Main object alignment (ablation) | EDA · ScanRefer VG-w/o-ON | Acc@0.25IoU: 21.8 -> 23.2 | Improved |
| Replace meaningful decoupled components with four randomly selected words (ablation) | EDA · ScanRefer | Acc@0.25IoU: 54.6 -> 52.0 | Hurt |
| Increase the weight of Main object to 1.0 while other component weights remain 0.2 (ablation) | EDA position-aligned loss · ScanRefer | Acc@0.25IoU: 54.6 -> 53.2 | Hurt |
| Use GT candidate boxes without retraining | EDA · ScanRefer Overall | Acc@0.25IoU: 54.59 -> 57.6 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- InstanceRefer — InstanceRefer: Cooperative Holistic Understanding for Visual Grounding on Point Clouds Through Instance Multi-Level Contextual Referring (ICCV 2021) · better on 2 of 2 shared results
- 3D-SPS — 3D-SPS: Single-Stage 3D Visual Grounding via Referred Point Progressive Selection (CVPR 2022) · better on 1 of 1 shared result
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
- G^3-LQ: Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding (CVPR 2024 · similarity 0.91)
- AugRefer: Advancing 3D Visual Grounding via Cross-Modal Augmentation and Spatial Relation-based Referring (AAAI 2025 · similarity 0.90)
- Look Around Before Locating: Considering Content and Structure Information for Visual Grounding (AAAI 2025 · similarity 0.90)
- DenseGrounding: Improving Dense Language-Vision Semantics for Ego-centric 3D Visual Grounding (ICLR 2025 · similarity 0.89)
- Towards CLIP-driven Language-free 3D Visual Grounding via 2D-3D Relational Enhancement and Consistency (CVPR 2024 · similarity 0.89)
- Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding (CVPR 2025 · similarity 0.89)
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.
