Abstract: 3D visual language multi-modal modeling plays an important role in actual human-computer interaction. However, the inaccessibility of large-scale 3D-language pairs restricts their ...
Abstract: Multimodal Large Language Models (MLLMs) have excelled in 2D image-text comprehension and image generation. Still, their understanding of the 3D world needs to be improved, limiting progress ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback