DeepView: View Synthesis With Learned Gradient Descent

John Flynn, Michael Broxton, Paul Debevec, Matthew DuVall, Graham Fyffe, Ryan Overbeck, Noah Snavely, Richard Tucker; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 2367-2376

Abstract


We present a novel approach to view synthesis using multiplane images (MPIs). Building on recent advances in learned gradient descent, our algorithm generates an MPI from a set of sparse camera viewpoints. The resulting method incorporates occlusion reasoning, improving performance on challenging scene features such as object boundaries, lighting reflections, thin structures, and scenes with high depth complexity. We show that our method achieves high-quality, state-of-the-art results on two datasets: the Kalantari light field dataset, and a new camera array dataset, Spaces, which we make publicly available.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Flynn_2019_CVPR,
author = {Flynn, John and Broxton, Michael and Debevec, Paul and DuVall, Matthew and Fyffe, Graham and Overbeck, Ryan and Snavely, Noah and Tucker, Richard},
title = {DeepView: View Synthesis With Learned Gradient Descent},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}