为了,我们的EmbodiedSAM采用先2D后3D的分割范式,对每一张RGB图像利用高效率的基础模型进行分割,得到初始的单帧分割结果,将此结果在3D空间进行可学习的优化和帧间融合,从而实现实时在线的3D Segment Anything。由于分割的任务交给了2D模型,我们的3D模型只需要充分利用时序和几何信息对来自2D的mask进行优化,以及对不同帧...
@misc{yang2023sam3d, title={SAM3D: Segment Anything in 3D Scenes}, author={Yunhan Yang, Xiaoyang Wu, Tong He, Hengshuang Zhao and Xihui Liu}, year={2023}, eprint={2306.03908}, archivePrefix={arXiv}, primaryClass={cs.CV} } Acknowledgements ...
TinySAM: Pushing the Envelope for Efficient Segment Anything Model TalkingFace DREAM-Talk: Diffusion-based Realistic Emotional Audio-driven Method for Single Image Talking Face Generation 3D场景编辑 Free-Editor: Zero-shot Text-driven 3D Scene Editing About Us NeRF相关 UniSDF: Unifying Neural Representa...