Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.04897 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917046852255744 |
|---|---|
| author | Wang, Tianxu Zhang, Zhuofan Zhu, Ziyu Fan, Yue Xiong, Jing Li, Pengxiang Ma, Xiaojian Li, Qing |
| author_facet | Wang, Tianxu Zhang, Zhuofan Zhu, Ziyu Fan, Yue Xiong, Jing Li, Pengxiang Ma, Xiaojian Li, Qing |
| contents | 3D visual grounding has made notable progress in localizing objects within complex 3D scenes. However, grounding referring expressions beyond objects in 3D scenes remains unexplored. In this paper, we introduce Anywhere3D-Bench, a holistic 3D visual grounding benchmark consisting of 2,886 referring expression-3D bounding box pairs spanning four different grounding levels: human-activity areas, unoccupied space beyond objects, individual objects in the scene, and fine-grained object parts. We assess a range of state-of-the-art 3D visual grounding methods alongside large language models (LLMs) and multimodal LLMs (MLLMs) on Anywhere3D-Bench. Experimental results reveal that space-level and part-level visual grounding pose the greatest challenges: space-level tasks require a more comprehensive spatial reasoning ability, for example, modeling distances and spatial relations within 3D space, while part-level tasks demand fine-grained perception of object composition. Even the best-performing models, Google Gemini-2.5-Pro and OpenAI o3, achieve just around 30% accuracy on space-level tasks and around 40% on part-level tasks, significantly lower than its performance on area-level and object-level tasks. These findings underscore a critical gap in current models' capacity to understand and reason about 3D scenes beyond object-level semantics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_04897 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes Wang, Tianxu Zhang, Zhuofan Zhu, Ziyu Fan, Yue Xiong, Jing Li, Pengxiang Ma, Xiaojian Li, Qing Computer Vision and Pattern Recognition 3D visual grounding has made notable progress in localizing objects within complex 3D scenes. However, grounding referring expressions beyond objects in 3D scenes remains unexplored. In this paper, we introduce Anywhere3D-Bench, a holistic 3D visual grounding benchmark consisting of 2,886 referring expression-3D bounding box pairs spanning four different grounding levels: human-activity areas, unoccupied space beyond objects, individual objects in the scene, and fine-grained object parts. We assess a range of state-of-the-art 3D visual grounding methods alongside large language models (LLMs) and multimodal LLMs (MLLMs) on Anywhere3D-Bench. Experimental results reveal that space-level and part-level visual grounding pose the greatest challenges: space-level tasks require a more comprehensive spatial reasoning ability, for example, modeling distances and spatial relations within 3D space, while part-level tasks demand fine-grained perception of object composition. Even the best-performing models, Google Gemini-2.5-Pro and OpenAI o3, achieve just around 30% accuracy on space-level tasks and around 40% on part-level tasks, significantly lower than its performance on area-level and object-level tasks. These findings underscore a critical gap in current models' capacity to understand and reason about 3D scenes beyond object-level semantics. |
| title | From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.04897 |