We present a technique for jointly denoising bursts of images taken from a handheld camera. In particular, we propose a convolutional neural network architecture for predicting spatially varying kernels that can both align and denoise frames, a synthetic data generation approach based on a realistic noise formation model, and an optimization guided by an annealed loss function to avoid undesirable local minima. Our model matches or outperforms the state-of-the-art across a wide range of noise levels on both real and synthetic data.
@article{arxiv.1712.02327,
title = {Burst Denoising with Kernel Prediction Networks},
author = {Ben Mildenhall and Jonathan T. Barron and Jiawen Chen and Dillon Sharlet and Ren Ng and Robert Carroll},
journal= {arXiv preprint arXiv:1712.02327},
year = {2018}
}
Comments
To appear in CVPR 2018 (spotlight). Project page: http://people.eecs.berkeley.edu/~bmild/kpn/