Research map / 3D vision / 3D generation
Towards Visual Query Localization in the 3D World
Liang Peng, Bohan Tan, Zhipeng Zhang, Haobo Li, Yifan Jiao, Xingping Dong, Libo 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 |
|---|---|---|---|
| add depth attention fusion (DAF) instead of element-wise addition (ablation) | LaF · 3DVQL (RGB-PC test set) | stAP: 0.007 -> 0.044 | Improved |
| add DAF (ablation) | LaF · 3DVQL (RGB-PC test set) | tAP: 0.134 -> 0.293 | Improved |
| add DAF (ablation) | LaF · 3DVQL (RGB-PC test set) | Success: 18.027 -> 46.041 | Improved |
| add DAF (ablation) | LaF · 3DVQL (RGB-PC test set) | stAP@0.05: 0.033 -> 0.222 | Improved |
| LaF fusion instead of best baseline fusion (PAF) | LaF vs PAF · 3DVQL (RGB-PC test set) | Success: 32.156 -> 46.041 | Improved |
| LaF instead of GAF | LaF vs GAF · 3DVQL (RGB-PC test set) | tAP: 0.291 -> 0.293 | No clear effect |
| add point cloud to RGB (PC-only instead of RGB-only) (ablation) | LaF · 3DVQL (supplementary modality settings) | Success: 12.302 -> 26.481 | Improved |
| use RGB-PC fusion instead of PC-D (ablation) | LaF · 3DVQL (supplementary modality settings) | Success: 30.632 -> 46.041 | Improved |
| use RGB-PC LaF instead of two-stage CenterPoint+MBPTracker | LaF vs TWO-Stage · 3DVQL (RGB-PC test set) | Success: 21.431 -> 46.041 | Improved |
| use LaF instead of point-tracking TAPIP3D | LaF vs TAPIP3D · 3DVQL (RGB-PC test set) | Success: 18.824 -> 46.041 | Improved |
| anchor-based head with 7 DoF GIoU loss (AF) instead of center-based head (ablation) | AF · 3DVQL (RGB-PC test set) | stAP@0.05: 0.015 | Hurt |
Closest papers by meaning
- RELOCATE: A Simple Training-Free Baseline for Visual Query Localization Using Region-Based Representations (CVPR 2025 · similarity 0.89)
- Localizing Events in Videos with Multimodal Queries (CVPR 2025 · similarity 0.88)
- GPVK-VL: Geometry-Preserving Virtual Keyframes for Visual Localization under Large Viewpoint Changes (CVPR 2025 · similarity 0.87)
- Learning from Videos for 3D World: Enhancing MLLMs with 3D Vision Geometry Priors (NeurIPS 2025 · similarity 0.86)
- Bridging the Gap between 2D and 3D Visual Question Answering: A Fusion Approach for 3D VQA (AAAI 2024 · similarity 0.86)
- Beyond 3D VQAs: Injecting 3D Spatial Priors into Vision-Language Models for Enhanced Geometric Reasoning (CVPR 2026 · similarity 0.86)
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.
