3 papers
cs.CV2026
OpenTrack3D: Towards Accurate and Generalizable Open-Vocabulary 3D Instance Segmentation
Zhishan Zhou, Siyuan Wei, Zengran Wang +3
Generalizing open-vocabulary 3D instance segmentation (OV-3DIS) to diverse, unstructured, and mesh-free environments is crucial for robotics and AR/VR, yet remains a significant ch…
cs.CV2026
MAG-3D: Multi-Agent Grounded Reasoning for 3D Understanding
Henry Zheng, Chenyue Fang, Rui Huang +3
Vision-language models (VLMs) have achieved strong performance in multimodal understanding and reasoning, yet grounded reasoning in 3D scenes remains underexplored. Effective 3D re…
cs.CV2026
Disc3D: Automatic Curation of High-Quality 3D Dialog Data via Discriminative Object Referring
Siyuan Wei, Chunjie Wang, Xiao Liu +3
3D Multi-modal Large Language Models (MLLMs) still lag behind their 2D peers, largely because large-scale, high-quality 3D scene-dialogue datasets remain scarce. Prior efforts hing…