Yan Zhang, Sergey Prokudin, Marko Mihajlovic, Qianli Ma, Siyu Tang
TL;DR通过神经网络学习隐式运动场,预测 3D 场景中新点的运动,以改善场景重建、运动跟踪和角色创建等应用。
Abstract
Understanding the dynamics of generic 3d scenes is fundamentally challenging
in computer vision, essential in enhancing applications related to scene
reconstruction, motion tracking, and avatar creation. In this