A group of scientists from Intel and the University of Illinois at Urbana–Champaign have published a paper called Learning to See in the Dark detailing a powerful machine-learning based image processing technique that allows regular cameras to take super-sharp pictures in very low light, without long exposures or the kinds of graininess associated with low-light photography.
The results are astounding.
We propose a new image processing pipeline that addresses the challenges of extreme low-light photography via
a data-driven approach. Specifically, we train deep neural
networks to learn the image processing pipeline for low-
light raw data, including color transformations, demosaic-
ing, noise reduction, and image enhancement. The pipeline
is trained end-to-end to avoid the noise amplification and
error accumulation that characterize traditional camera pro-
cessing pipelines in this regime.
Learning to See in the Dark [Chen Chen, Qifeng Chen, Jia Xu and Vladlen Koltun/Arxiv]
Learning to See in the Dark [Chen Chen, Qifeng Chen, Jia Xu and Vladlen Koltun/University of Illinois]
(via Four Short Links)