Colorization

Let there be Color! (SIGGRAPH 2016)

Let there be Color!: Joint End-to-end Learning of Global and Local Image Priors for Automatic Image Colorization with Simultaneous Classification
Project | Online demo | Torch code

Style2Paints

Project | github repo: server-client (tensorflow + keras)

Style2Paints v1 (ACPR 2017)

Style Transfer for Anime Sketches with Enhanced Residual U-net and Auxiliary Classifier GAN

• A feed-forward network to apply the style of a painting to a sketch.
• An enhanced residual U-net capable of handling paired images with unbalanced information quantity.
• An effective way to train residual U-net with two additional loss.
• A discriminator modified from AC-GAN suitable to deal with paintings of different style.

Style2Paints v3 (SIGGRAPH ASIA 2018)

Two-stage Sketch Colorization by CUHK