1 paper · 1 filter
Liu He, Xiao Zeng, Yizhi Song +7
Multimodal Large Language Models (MLLMs) struggle with accurately capturing camera-object relations, especially for object orientation, camera viewpoint, and camera shots. This ste…