OpenEMMA, a permissively licensed open source "reproduction" of Waymo’s EMMA model.

Unsupervised Hierarchical Semantic Segmentation with Multiview Cosegmentation and Clustering Transformers

Supercharge Your Model Training

Universal Weakly Supervised Segmentation by Pixel-to-Segment Contrastive Learning

Repository for "Space-Time Correspondence as a Contrastive Random Walk" (NeurIPS 2020)

PySlowFast: video understanding codebase from FAIR for reproducing state-of-the-art video models.

Adversarial Structure Matching for Structured Prediction Tasks

SegSort: Segmentation by Discriminative Sorting of Segments

disentanglement_lib is an open-source library for research on learning disentangled representations.

Models and examples built with TensorFlow

Adaptive Affinity Fields for Semantic Segmentation

Improving Generalization via Scalable Neighborhood Component Analysis

Unsupervised Feature Learning via Non-parametric Instance Discrimination

A full-fledged yet minimalistic CUDA-based convolutional neural network library from scratch in C++