3D Human
3D human modeling focuses on creating realistic three-dimensional representations of humans, aiming for accurate geometry, texture, and dynamic movement. Current research emphasizes generating high-fidelity models from limited input, such as single images or sparse video views, often employing neural rendering techniques like Gaussian splatting and neural radiance fields, along with diffusion models and graph convolutional networks for pose estimation and animation. This field is crucial for advancements in virtual reality, animation, human-computer interaction, and other applications requiring accurate and realistic human representations.
Papers
HumanGaussian: Text-Driven 3D Human Generation with Gaussian Splatting
Xian Liu, Xiaohang Zhan, Jiaxiang Tang, Ying Shan, Gang Zeng, Dahua Lin, Xihui Liu, Ziwei Liu
HumanRef: Single Image to 3D Human Generation via Reference-Guided Diffusion
Jingbo Zhang, Xiaoyu Li, Qi Zhang, Yanpei Cao, Ying Shan, Jing Liao