PyTorch code and models for V-JEPA self-supervised learning from video.

Paper code for "Generating Description for Sequential Images with Local-Object Attention Conditioned on Global Semantic Context"

A research source code for co-transduction on retrieval

The pototype of patent and e-commerce comondity image search

Comes from the paper"Noise-contrastive estimation based on relative neighbour sampling"

TLS/SSL and crypto library

Repo for counting stars and contributing. Press F to pay respect to glorious developers.

Unsupervised Feature Learning via Non-parametric Instance Discrimination

Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow