Abstract: Visual affordance grounding aims to segment all possible interaction regions between people and objects from an image/video, which benefits many applications, such as robot grasping and ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results