1 paper
Zhengqing Wang, Yuefan Wu, Jiacheng Chen +2
This paper proposes a neural rendering approach that represents a scene as "compressed light-field tokens (CLiFTs)", retaining rich appearance and geometric information of a scene.…