Sequence Modeling with Structured State Spaces
A U-Net combined with a variational auto-encoder that is able to learn conditional distributions over semantic segmentations.
Code for KDD'20 "Generative Pre-Training of Graph Neural Networks"
Code for Parameter Prediction for Unseen Deep Architectures (NeurIPS 2021)
Tensorflow 2.x based implementation of EDSR, WDSR and SRGAN for single image super-resolution
Grounded Language-Image Pre-training
PyTorch implementation of paper "Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes"
[CVPR2019] The Perfect Match: 3D Point Cloud Matching with Smoothed Densities