Codes

The following are some public codes related to our research work:

  • Generative View Synthesis from 2D Semantics - [Code(github)]
  • Self-supervised Single-view 3D Reconstruction via Semantic Consistency - [Code(github)]
  • DeepGMR: Learning Latent Gaussian Mixture Models for Registration - [Code(github)]
  • Appearance Consensus Driven Self-Supervised Human Mesh Recovery - [Code(github)]
  • Two-shot Spatially-varying BRDF and Shape Estimation - [Code(github)]
  • Self-supervised viewpoint learning from image collections - [Code(github)]
  • Self-supervised point cloud reconstruction networks - [Code(github)]
  • SENSE: a Shared Encoder Network for Scene-flow Estimation - [Code(github)]
  • Pixel-Adaptive Convolutional Neural Networks - [Code(github)]
  • SCOPS: Self-supervised Co-part Segmentation - [Code(github)]
  • Gated Shape CNNs for Semantic Segmentation - [Code(github)]
  • Competitive Collaboration - [Code(github)]
  • Superpixel Sampling Networks - [Code(github)]
  • SPLATNet for 2D-3D semantic segmentation - [Code(github)]
  • Superpixels trained with segmentation aware affinity loss (SEAL) - [Code(github)]
  • Video Propagation Networks - [Project Page] [Code(github)]
  • Semantic Video CNNs through Representation Warping - [Project Page] [Code(github)]
  • Bilateral Inception Networks - [Project Page] [Code(github)]
  • Learning Sparse High Dimensional Filters - [Project Page] [Code(github)]
  • Semantic Optical Flow - [Project Page] [Code(zip)]
  • Efficient Facade Segmentation using Auto-Context - [Project Page] [Code(bitbucket)]