Previous multimodal visual-tactile image representation learning (VTL) methods have achieved significant success in object understanding through large-scale training data. However, obtaining sufficient training data is often infeasible, and the above methods s... ...

请注册登录后继续浏览