Jun-Yan Zhu

Image-to-Image Translation in PyTorch

This repo provides PyTorch implementations of CycleGAN and pix2pix, two AI models that transform images from one visual style to another, like turning horses into zebras or rough sketches into realistic photos.

Software that can generate photos from paintings, turn horses into zebras, perform style transfer, and more.

A research implementation of CycleGAN, an AI technique that converts images from one visual style to another, like photos to Monet paintings or horses to zebras, without needing matched image pairs.

Interactive Image Generation via Generative Adversarial Networks

Deep learning software for colorizing black and white images with a few clicks.

Toward Multimodal Image-to-Image Translation

Cool vision, learning, and graphics papers on Cats!

[NeurIPS 2018] Visual Object Networks: Image Generation with Disentangled 3D Representation.

Light field video applications (e.g. video refocusing, focus tracking, changing aperture and view)

A simple 3D face alignment and warping demo.

code for predicting and improving visual realism in composite images

a program for selecting attractive/serious portraits from a personal photo collection

a boosting-based Multiple Instance Learning (MIL) package that includes MIL-Boost and MCIL-Boost

An expression training App that helps users mimic their own expressions.

No description.