4 Dimensional Generation
4D generation aims to create dynamic 3D models that evolve over time, focusing on generating realistic and temporally consistent animations. Current research heavily utilizes diffusion models, often incorporating multi-view approaches and physics-based simulations to improve realism and efficiency, with some methods employing animatable meshes or Gaussian splatting for representation. This field is significant for its potential applications in animation, virtual reality, and autonomous driving, offering advancements in realistic scene simulation and content creation.
Papers
Trans4D: Realistic Geometry-Aware Transition for Compositional Text-to-4D Synthesis
Bohan Zeng, Ling Yang, Siyu Li, Jiaming Liu, Zixiang Zhang, Juanxi Tian, Kaixin Zhu, Yongzhen Guo, Fu-Yun Wang, Minkai Xu, Stefano Ermon, Wentao Zhang
DreamMesh4D: Video-to-4D Generation with Sparse-Controlled Gaussian-Mesh Hybrid Representation
Zhiqi Li, Yiming Chen, Peidong Liu