• Publications
  • Influence
Striving for Simplicity: The All Convolutional Net
TLDR
We propose a new architecture that consists solely of convolutional layers and yields competitive or state of the art performance on several object recognition datasets (CIFAR-10, CIFAR100, ImageNet). Expand
  • 2,372
  • 273
  • PDF
A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation
  • N. Mayer, Eddy Ilg, +4 authors T. Brox
  • Computer Science, Mathematics
  • IEEE Conference on Computer Vision and Pattern…
  • 7 December 2015
TLDR
We propose three synthetic stereo video datasets with sufficient realism, variation, and size to successfully train large networks for scene flow estimation. Expand
  • 1,037
  • 253
  • PDF
FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks
The FlowNet demonstrated that optical flow estimation can be cast as a learning problem. However, the state of the art with regard to the quality of the flow has still been defined by traditionalExpand
  • 1,364
  • 202
  • PDF
CARLA: An Open Urban Driving Simulator
TLDR
We introduce CARLA, an open-source simulator for autonomous driving research. Expand
  • 944
  • 195
  • PDF
Generating Images with Perceptual Similarity Metrics based on Deep Networks
TLDR
We propose a class of loss functions, which we call deep perceptual similarity metrics (DeePSiM), that mitigate this problem. Expand
  • 654
  • 52
  • PDF
DeMoN: Depth and Motion Network for Learning Monocular Stereo
TLDR
We train a convolutional network end-to-end to compute depth and camera motion from successive, unconstrained image pairs. Expand
  • 379
  • 47
  • PDF
On Evaluation of Embodied Navigation Agents
TLDR
We convened a working group to study empirical methodology in navigation research. Expand
  • 174
  • 43
  • PDF
Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space
TLDR
Generating high-resolution, photo-realistic images has been a long-standing goal in machine learning. Expand
  • 408
  • 40
  • PDF
Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks
TLDR
We show that the feature representation learned by the Exemplar-CNN performs well on two very different tasks: object classification and descriptor matching. Expand
  • 275
  • 39
  • PDF
Discriminative Unsupervised Feature Learning with Convolutional Neural Networks
TLDR
We introduce a novel training procedure for training a convolutional neural network using only unlabeled data that does not require any labeled data. Expand
  • 463
  • 38
  • PDF