C
Christian Theobalt
Researcher at Max Planck Society
Publications - 508
Citations - 34680
Christian Theobalt is an academic researcher from Max Planck Society. The author has contributed to research in topics: Motion capture & Computer science. The author has an hindex of 89, co-authored 450 publications receiving 25487 citations. Previous affiliations of Christian Theobalt include Stanford University & Facebook.
Papers
More filters
Proceedings ArticleDOI
Text-Based Motion Synthesis with a Hierarchical Two-Stream RNN
TL;DR: The authors proposed a hierarchical two-stream sequential model to explore a finer joint-level mapping between natural language sentences and the corresponding 3D pose sequences of the motions, achieving state-of-the-art performance on the KIT Motion-Language Dataset.
Journal ArticleDOI
F2-NeRF: Fast Neural Radiance Field Training with Free Camera Trajectories
Peng Wang,Yuan Liu,Zhaoxi Chen,Lingjie Liu,Ziwei Liu,Taku Komura,Christian Theobalt,Wenping Wang +7 more
TL;DR: In this paper , a grid-based view synthesis framework called F2-NERF (Fast-Free-NeRF) is proposed for novel view synthesis, which enables arbitrary input camera trajectories and only costs a few minutes for training.
Posted Content
Video Depth-From-Defocus
TL;DR: The core algorithmic ingredient is a new video-based depth-from-defocus algorithm that computes space-time-coherent depth maps, deblurred all-in-focus video, and the focus distance for each frame.
Journal ArticleDOI
Fast Gravitational Approach for Rigid Point Set Registration With Ordinary Differential Equations
TL;DR: Fast Gravitational Approach (FGA) as mentioned in this paper models the source and target point sets as rigid particle swarms with masses interacting in a globally multiply-linked manner while moving in a simulated gravitational force field.
Journal ArticleDOI
Grid-guided Neural Radiance Fields for Large Urban Scenes
Linning Xu,Yuanbo Xiangli,Sida Peng,Xingang Pan,Nanxuan Zhao,Christian Theobalt,B. Z. Dai,Dahua Lin +7 more
TL;DR: In this article , the authors propose to use a compact multiresolution ground feature plane representation to coarsely capture the scene, and complement it with positional encoding inputs through another NeRF branch for rendering in a joint learning fashion.