At this week’s SIGGRAPH conference, researchers from MIT and Google presented a new deep learning system that automatically retouches your photos like a professional photographer in real-time, before you even take them.
Using a TITAN X GPU and the cuDNN-accelerated TensorFlow deep learning framework, the researchers trained their neural network on a dataset of 5,000 photographs that had been retouched by different photographers. The algorithms were able to learn what sort of improvements to make to different photos, such as adjusting the brightness or saturation.
According to their paper, the breakthrough achieved was slimming down the algorithms so they are able to processes high-resolution images on a smartphone in milliseconds, providing a real-time viewfinder at 1080p resolution, while still capturing high-frequency effects.
“This technology has the potential to be very useful for real-time image enhancement on mobile platforms,” says Jon Barron of Google Research. “Using machine learning for computational photography is an exciting prospect but is limited by the severe computational and power constraints of mobile phones. This paper may provide us with a way to sidestep these issues and produce new, compelling, real-time photographic experiences without draining your battery or giving you a laggy viewfinder experience.”