1 paper · 1 filter
Shengqiong Wu, Lanhu Wu, Mingyang Bao +5
Recent advances in large vision models (LVMs) have shifted from modality-specific designs toward unified architectures that jointly process images, videos, and 3D data. However, ex…