LLaVA-3D could perform both 2D and 3D vision-language tasks. The left block (b) shows that compared with previous 3D LMMs, our LLaVA-3D achieves state-of-the-art performance across a wide range of 3D ...
Abstract: Augmented Reality (AR) integration via handheld devices is a subject of significant interest for mobile applications and interactions between humans and machines. The mobile AR technique ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results