From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes
This addresses a critical gap in AI's ability to understand 3D scenes comprehensively, which is important for applications like robotics and augmented reality, though it is incremental as it builds on existing 3D visual grounding work.
The paper tackles the problem of 3D visual grounding beyond just objects by introducing Anywhere3D-Bench, a benchmark with 2,886 expression-bounding box pairs across four levels including human-activity areas, unoccupied space, objects, and object parts. Results show that even top models like Google Gemini-2.5-Pro and OpenAI o3 achieve only around 30% accuracy on space-level tasks and 40% on part-level tasks, highlighting significant performance gaps.
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.