• MentorMix - Robust deep learning on noisy labelled data.
  • Future Prediction in Video - First and currently the best model for joint path and activity prediction.
  • Text Image Residual Gating - New method of using vision and language research for retrieval.
  • Eidetic-3D LSTM - State-of-the-art of self-supervised video learning model.
  • MentorNet - Weakly-supervised deep learning method.
  • FVTA - Focal Visual-Text Attention for Visual Question Answering.
  • Graph Distillation - Code for ECCV paper on knowledge distillation for action detection.
  • SURREAL - Open-source, reproducible, scalable distributed reinforcement learning framework.
  • Self-paced Learning - An implementation for self-paced learning used in our paper.
  • Learning2Tile - Learning "an optimal" spatial BoW representation from data.

Data Set