spatial relations
The contextual relationships and arrangement of objects in space, critical for understanding scenes in computer vision and enabling reasoning about physical interactions between objects.
- From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes
- IR3D-Bench: Evaluating Vision-Language Model Scene Understanding as Agentic Inverse Rendering
- Part-Aware Bottom-Up Group Reasoning for Fine-Grained Social Interaction Detection
- RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics
- SceneForge: Enhancing 3D-text alignment with Structured Scene Compositions
- See&Trek: Training-Free Spatial Prompting for Multimodal Large Language Model