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
Posted Content
Semi-Supervised Disparity Estimation with Deep Feature Reconstruction
TL;DR: In this paper, a semi-supervised pipeline was proposed to adapt DispNet to a real-world domain by joint supervised training on labeled synthetic data and self-vised training on unlabeled real data.
Posted Content
Group Pruning using a Bounded-Lp norm for Group Gating and Regularization
TL;DR: In this paper, a gating factor after every convolutional layer is proposed to induce channel level sparsity, encouraging insignificant channels to become exactly zero, and a bounded variant of the L1 regularizer interpolates between L1 and L0-norms to retain performance of the network at higher pruning rates.
Posted Content
CrossCLR: Cross-modal Contrastive Learning For Multi-modal Video Representations
TL;DR: This article proposed a contrastive loss to learn cross-modal embeddings by contrasting positive pairs from sets of negative samples and exclude highly related samples from the negative samples to avoid false negatives.
Posted Content
Unsupervised Object Learning via Common Fate
Matthias Tangemann,Steffen Schneider,Julius von Kügelgen,Francesco Locatello,Peter V. Gehler,Thomas Brox,Matthias Kümmerer,Matthias Bethge,Bernhard Schölkopf +8 more
TL;DR: In this paper, a generative model is trained on the masks of the background and the moving objects, respectively, and background and foreground models are combined in a conditional dead leaves scene model to sample novel scene configurations where occlusions and depth layering arise naturally.
Book ChapterDOI
Segmentation in Point Clouds from RGB-D Using Spectral Graph Reduction
Margret Keuper,Thomas Brox +1 more
TL;DR: This chapter combines a state-of-the-art method for natural RGB image segmentation based on spectral graph analysis with an RGB-D boundary detector and shows that spectral graph reduction can be employed in this case, facilitating the computation ofRGB-D segmentations in large datasets.