Research map / Vision-language and multimodal / CLIP and zero-shot
UZ3DVG: Unaided Zero-Shot 3D Visual Grounding with Generated Language Conditions
Wenbin Tan, Jiawen Lin, Yuan Xie, Yachao Zhang, Yanyun Qu
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 TSVDE (text-guided sampling) over random sampling (ablation) | UZ3DVG · ScanRefer val | Acc@0.25: 40.75 -> 41.83 | Improved |
| add Reasoning Chain Distillation (RCD) (ablation) | UZ3DVG with TSVDE and GeoSM · ScanRefer val | Acc@0.25: 43.15 -> 45.42 | Improved |
| add GeoSM on top of TSVDE (ablation) | UZ3DVG with TSVDE and RCD · ScanRefer val | Acc@0.25: 43.26 -> 45.42 | Improved |
| all three modules combined vs none (ablation) | UZ3DVG · ScanRefer val | Acc@0.25: 40.75 -> 45.42 | Improved |
| generate pseudo-labels with Doubao-seed-1.6 instead of Qwen3-VL-Plus (ablation) | UZ3DVG · ScanRefer val | Acc@0.25: 43.87 -> 45.42 | Improved |
| increase pseudo-labels from 10K to 20K (ablation) | UZ3DVG · ScanRefer val | Acc@0.25: 25.71 -> 36.88 | Improved |
| increase pseudo-labels from 30K to 40K (ablation) | UZ3DVG · ScanRefer val | Acc@0.25: 43.36 -> 45.42 | Improved |
| project 2D boxes into 3D with MSAG-RC instead of 2D-only descriptions (ablation) | UZ3DVG · ScanRefer val | Acc@0.50: 22.83 -> 39.47 | Improved |
| add reasoning chains to MSAG-RC descriptions (ablation) | UZ3DVG · ScanRefer val | Acc@0.50: 39.47 -> 41.08 | Improved |
| use UZ3DVG instead of ZSVG3D without Mask3D | zero-shot 3DVG · ScanRefer val (non-Mask3D) | Acc@0.25: 20.00 -> 43.13 | Improved |
| UZ3DVG vs SeeGround on Nr3D transfer | zero-shot 3DVG · Nr3D val | accuracy: 46.1 -> 46.5 | Improved |
| UZ3DVG vs SeeGround on Nr3D Dep. subset | zero-shot 3DVG · Nr3D val (View-Dependent) | accuracy: 42.3 -> 39.3 | Hurt |
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.
- SeeGround — SeeGround: See and Ground for Zero-Shot Open-Vocabulary 3D Visual Grounding (CVPR 2025) · 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
- Visual Programming for Zero-shot Open-Vocabulary 3D Visual Grounding (CVPR 2024 · similarity 0.94)
- SPAZER: Spatial-Semantic Progressive Reasoning Agent for Zero-shot 3D Visual Grounding (NeurIPS 2025 · similarity 0.92)
- SeeGround: See and Ground for Zero-Shot Open-Vocabulary 3D Visual Grounding (CVPR 2025 · similarity 0.92)
- Chain of Semantics Programming in 3D Gaussian Splatting Representation for 3D Vision Grounding (CVPR 2025 · similarity 0.92)
- Grounded 3D-Aware Spatial Vision-Language Modeling (CVPR 2026 · similarity 0.91)
- S$^2$-MLLM: Boosting Spatial Reasoning Capability of MLLMs for 3D Visual Grounding with Structural Guidance (CVPR 2026 · similarity 0.90)
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.
