• Publications
  • Influence
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images
TLDR
We propose an end-to-end deep learning architecture that produces a 3D shape in triangular mesh from a single color image, leveraging perceptual features extracted from the input image. Expand
  • 429
  • 89
  • PDF
Interactive Image Segmentation with Latent Diversity
TLDR
We present an end-to-end learning approach to interactive image segmentation that tackles the multimodality problem head-on. Expand
  • 62
  • 19
  • PDF
What Do Single-View 3D Reconstruction Networks Learn?
TLDR
We demonstrate that the performance of modern convolutional networks for single-view object reconstruction can be surpassed even without explicitly inferring the 3D structure of objects. Expand
  • 103
  • 16
  • PDF
Combinatorial Optimization with Graph Convolutional Networks and Guided Tree Search
TLDR
We present a learning-based approach to computing solutions for certain NP-hard problems that can be expressed in terms of graphs. Expand
  • 147
  • 11
  • PDF
PointPWC-Net: A Coarse-to-Fine Network for Supervised and Self-Supervised Scene Flow Estimation on 3D Point Clouds
TLDR
We propose a novel end-to-end deep scene flow model, called PointPWC-Net, on 3D point clouds in a coarse- to-fine fashion. Expand
  • 24
  • 11
  • PDF
Simultaneous video defogging and stereo reconstruction
TLDR
We present a method to jointly estimate scene depth and recover the clear latent image from a foggy video sequence. Expand
  • 68
  • 9
  • PDF
Perspective Motion Segmentation via Collaborative Clustering
TLDR
This paper addresses real-world challenges in the motion segmentation problem, including perspective effects, missing data, and unknown number of motions. Expand
  • 51
  • 7
  • PDF
Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformation
TLDR
We study the problem of shape generation in 3D mesh representation from a few color images with known camera poses. Expand
  • 53
  • 6
  • PDF
Video Co-segmentation for Meaningful Action Extraction
TLDR
We develop an analogous video co-segmentation framework for common action extraction, which allows us to extract the desired action without having to use higher level learning. Expand
  • 44
  • 6
  • PDF
Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness
TLDR
We present Uncertainty-aware Cascaded Stereo Network (UCS-Net) for 3D reconstruction from multiple RGB images. Expand
  • 20
  • 5
  • PDF
...
1
2
3
...