• Publications
  • Influence
The Cityscapes Dataset for Semantic Urban Scene Understanding
TLDR
We introduce Cityscapes, a benchmark suite and large-scale dataset to train and test approaches for pixel-level and instance-level semantic labeling. Expand
  • 3,845
  • 1005
  • PDF
A Database and Evaluation Methodology for Optical Flow
TLDR
The quantitative evaluation of optical flow algorithms by Barron et al. (1994) led to significant advances in performance. Expand
  • 2,145
  • 309
  • PDF
Playing for Data: Ground Truth from Computer Games
TLDR
We present an approach to rapidly creating pixel-accurate semantic label maps for images extracted from modern computer games, with no access to the source code or the content. Expand
  • 749
  • 158
  • PDF
Fields of Experts: a framework for learning image priors
TLDR
We develop a framework for learning generic, expressive image priors that capture the statistics of natural scenes and can be used for a variety of machine vision tasks. Expand
  • 990
  • 96
  • PDF
Fields of Experts
TLDR
We develop a framework for learning generic, expressive image priors that capture the statistics of natural scenes and can be used for a variety of machine vision tasks. Expand
  • 625
  • 81
  • PDF
Shrinkage Fields for Effective Image Restoration
  • U. Schmidt, S. Roth
  • Computer Science
  • IEEE Conference on Computer Vision and Pattern…
  • 23 June 2014
TLDR
We propose shrinkage fields, a random field-based architecture that combines the image model and the optimization algorithm in a single unit. Expand
  • 359
  • 73
  • PDF
Continuous Energy Minimization for Multitarget Tracking
TLDR
In this work, we propose an alternative formulation of multitarget tracking as minimization of a continuous energy that corresponds to a more complete representation of the problem. Expand
  • 489
  • 65
  • PDF
Monocular 3D pose estimation and tracking by detection
TLDR
Automatic recovery of 3D human pose from monocular image sequences is a challenging and important research topic with numerous applications. Expand
  • 513
  • 56
  • PDF
Pictorial structures revisited: People detection and articulated pose estimation
TLDR
This paper proposes a generic approach based on the pictorial structures framework for non-rigid object detection and articulated pose estimation that outperforms the state of the art on three recently proposed datasets. Expand
  • 480
  • 55
Benchmarking Denoising Algorithms with Real Photographs
  • Tobias Plotz, S. Roth
  • Computer Science
  • IEEE Conference on Computer Vision and Pattern…
  • 5 July 2017
TLDR
We develop a methodology for benchmarking denoising techniques on real photographs by means of real photographs. Expand
  • 195
  • 53
  • PDF