Machine Learning for Art
ml4a is a collection of tools and educational resources which apply techniques from machine learning to arts and creativity.

Salient object detection
Extract a segmentation mask of an image's foreground.



DeepDream
Visualize and compose abstract textures using neural network feature detectors.



Neural style transfer
Generate images with the style characteristics of other images.



Image-to-image translation
Generate photorealistic imagery from semantic label maps.





Semantic segmentation
Segment the pixels of an image into their respective object categories.




Reversible generative models
Lossless inversion of face images into the latent space of a generative model.





Lip-syncing videos
Animate mouths in images and videos, lip-syncing from audio recordings.




Optical flow
Compute optical flow between images to detect and measure motion.

