D
Daniel Cremers
Researcher at Technische Universität München
Publications - 702
Citations - 55592
Daniel Cremers is an academic researcher from Technische Universität München. The author has contributed to research in topics: Image segmentation & Computer science. The author has an hindex of 99, co-authored 655 publications receiving 44957 citations. Previous affiliations of Daniel Cremers include Siemens & University of Mannheim.
Papers
More filters
Journal ArticleDOI
A Multiphase Dynamic Labeling Model for Variational Recognition-driven Image Segmentation
TL;DR: This framework fundamentally extends previous work on shape priors in level set segmentation by directly addressing the central question of where to apply which prior and may selectively use specific shape knowledge for simultaneously enhancing segmentation and recognizing shape.
Proceedings ArticleDOI
A primal-dual framework for real-time dense RGB-D scene flow
TL;DR: This paper presents the first method to compute dense scene flow in real-time for RGB-D cameras, based on a variational formulation where brightness constancy and geometric consistency are imposed, and which is able to estimate heterogeneous and non-rigid motions at a high frame rate.
Proceedings ArticleDOI
Direct Sparse Visual-Inertial Odometry using Dynamic Marginalization.
TL;DR: In this paper, the authors proposed a novel approach for visual-inertial odometry, which jointly estimates camera poses and sparse scene geometry by minimizing photometric and IMU measurement errors in a combined energy functional.
Book ChapterDOI
Video Super Resolution Using Duality Based TV-L1 Optical Flow
TL;DR: This paper employs a recently proposed quadratic relaxation scheme for high accuracy optic flow estimation and estimates a high resolution image using a variational approach that models the image formation process and imposes a total variation regularity of the estimated intensity map.
Proceedings ArticleDOI
Learning by Association — A Versatile Semi-Supervised Training Method for Neural Networks
TL;DR: This work proposes a new framework for semi-supervised training of deep neural networks inspired by learning in humans and demonstrates the capabilities of learning by association on several data sets and shows that it can improve performance on classification tasks tremendously by making use of additionally available unlabeled data.