T
Thomas Brox
Researcher at University of Freiburg
Publications - 353
Citations - 127470
Thomas Brox is an academic researcher from University of Freiburg. The author has contributed to research in topics: Segmentation & Optical flow. The author has an hindex of 99, co-authored 329 publications receiving 94431 citations. Previous affiliations of Thomas Brox include Dresden University of Technology & University of California, Berkeley.
Papers
More filters
Proceedings ArticleDOI
Unsupervised Generation of a View Point Annotated Car Dataset from Videos
Nima Sedaghat,Thomas Brox +1 more
TL;DR: This paper presents an approach that creates a dataset of images annotated with bounding boxes and viewpoint labels in a fully automated manner from videos and shows that it can expand a large dataset of annotated single images and obtain improved performance when training a viewpoint regressor on this joined dataset.
Book ChapterDOI
WarpCut: fast obstacle segmentation in monocular video
TL;DR: The proposed technique for fast segmentation of stationary obstacles from video recorded by a single camera that is installed in a moving vehicle leads to fast and accurate obstacle segmentation and distance estimation without prior knowledge about the size, shape or base point of obstacles.
Book ChapterDOI
Topometric Localization with Deep Learning
TL;DR: In this paper, a vision-based localization approach that learns from LiDAR-based methods by using their output as training data is proposed, which combines a cheap, passive sensor with an accuracy that is on-par with LBS.
Multi-Stage Region Merging for Image Segmentation
Thomas Brox,Dirk Farin +1 more
TL;DR: A novel algorithm is developed, which is a multi-stage generalization of conventional region merging, which uses a sequence of different criteria to achieve a semantically and subjectively superior segmentation result.
Nonparametric Density Estimation for Human Pose Tracking
Thomas Brox,Bodo Rosenhahn,Uwe G. Kersting,Daniel Cremers,Katrin Franke,Klaus-Robert Müller,Bertram Nickolay,Ralf Schäfer +7 more
TL;DR: In this paper, the supplement of prior knowledge about joint angle configurations in the scope of 3D human pose tracking is considered for a nonparametric Parzen density estimation in the 12-dimensional joint configuration space.