Research map / 3D vision / 3D generation
AnyGround3D: Towards Grounding Any 3D Object in the Wild via 2D-to-3D Lifting
Yingping Liang, Wenxuan Guo
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 2D-to-3D distillation encoder pre-training (ablation) | AnyGround3D · 3EED Union | Acc@0.25: 66.34 -> 66.76 | Improved |
| add 2D-to-3D distillation encoder pre-training (ablation) | AnyGround3D · 3EED Union | Acc@0.5: 31.31 -> 31.34 | No clear effect |
| add VLM caption synthesis (ablation) | AnyGround3D · 3EED Union | Acc@0.25: 64.10 -> 66.76 | Improved |
| add VLM caption synthesis (ablation) | AnyGround3D · 3EED Union | Acc@0.5: 28.59 -> 31.34 | Improved |
| add generated 2D-lifted 3D data (ablation) | AnyGround3D · 3EED Union | Acc@0.25: 65.21 -> 66.76 | Improved |
| add generated 2D-lifted 3D data (ablation) | AnyGround3D · 3EED Union | Acc@0.5: 31.11 -> 31.34 | Improved |
| use AnyGround3D instead of 3EED baseline on quadruped | 3EED benchmark · 3EED Quadruped | Acc@0.25: 53.31 -> 61.60 | Improved |
| use AnyGround3D instead of 3EED baseline on vehicle | 3EED benchmark · 3EED Vehicle | Acc@0.25: 80.86 -> 80.88 | No clear effect |
| use AnyGround3D instead of 3EED baseline on drone | 3EED benchmark · 3EED Drone | Acc@0.5: 9.75 -> 9.81 | No clear effect |
| train on lifted data instead of 3EED data, evaluate zero-shot | Mono3DRefer · Mono3DRefer val | Acc@0.5: 12.19 -> 49.59 | Improved |
| use AnyGround3D instead of WildRefer on WildRefer union | WildRefer benchmark · WildRefer Union | Acc@0.5: 36.70 -> 39.00 | Improved |
Closest papers by meaning
- From Thousands to Billions: 3D Visual Language Grounding via Render-Supervised Distillation from 2D VLMs (ICML 2025 · similarity 0.90)
- From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes (NeurIPS 2025 · similarity 0.89)
- SeeGround: See and Ground for Zero-Shot Open-Vocabulary 3D Visual Grounding (CVPR 2025 · similarity 0.89)
- ReasonGrounder: LVLM-Guided Hierarchical Feature Splatting for Open-Vocabulary 3D Visual Grounding and Reasoning (CVPR 2025 · similarity 0.88)
- Towards CLIP-driven Language-free 3D Visual Grounding via 2D-3D Relational Enhancement and Consistency (CVPR 2024 · similarity 0.88)
- G^3-LQ: Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding (CVPR 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.
