• Publications
  • Influence
Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
tl;dr
We present a new method that addresses this task by employing two deep network stacks: one that makes a coarse global prediction based on the entire image, and another that refines this prediction locally. Expand
  • 1,568
  • 345
  • Open Access
OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
tl;dr
We present an integrated framework for using Convolutional Networks for classification, localization and detection with a single ConvNet. Expand
  • 3,604
  • 295
  • Open Access
Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture
  • D. Eigen, R. Fergus
  • Computer Science
  • IEEE International Conference on Computer Vision…
  • 17 November 2014
tl;dr
In this paper we address three different computer vision tasks using a single basic architecture: depth prediction, surface normal estimation, and semantic labeling. Expand
  • 1,416
  • 187
  • Open Access
Restoring an Image Taken through a Window Covered with Dirt or Rain
tl;dr
We present a post-capture image processing solution that can remove localized rain and dirt artifacts from a single image. Expand
  • 255
  • 29
  • Open Access
Nonparametric image parsing using adaptive neighbor sets
  • D. Eigen, R. Fergus
  • Computer Science
  • IEEE Conference on Computer Vision and Pattern…
  • 16 June 2012
tl;dr
This paper proposes a non-parametric approach to scene parsing inspired by the work of Tighe and Lazebnik [22]. Expand
  • 78
  • 14
  • Open Access
Learning Factored Representations in a Deep Mixture of Experts
tl;dr
Mixtures of Experts combine the outputs of several "expert" networks, each of which specializes in a different part of the input space. Expand
  • 81
  • 10
  • Open Access
End-to-end integration of a Convolutional Network, Deformable Parts Model and non-maximum suppression
tl;dr
We propose an object detection system that integrates a Convolutional Network, Deformable Parts model and NMS loss in an end-to-end fashion. Expand
  • 77
  • 10
  • Open Access
Finding Task-Relevant Features for Few-Shot Learning by Category Traversal
tl;dr
We introduce a Category Traversal Module that can be inserted as a plug-and-play module into most metric-learning based few-shot learners. Expand
  • 62
  • 8
  • Open Access
Unsupervised Learning of Spatiotemporally Coherent Metrics
tl;dr
We study unsupervised feature learning with convolutional networks in the context of temporally coherent unlabeled data, using the assumption that adjacent video frames contain semantically similar information. Expand
  • 88
  • 6
  • Open Access
Understanding Deep Architectures using a Recursive Convolutional Network
tl;dr
A key challenge in designing convolutional network models is sizing them appropriately. Expand
  • 108
  • 5
  • Open Access