End-to-End Video Compressive Sensing Using Anderson-Accelerated Unrolled Networks

• Published in: IEEE International Conference on Computational Photography pp. 1 - 12
• Main Authors: Li, Yuqi, Qi, Miao, Gulve, Rahul, Wei, Mian, Genov, Roman, Kutulakos, Kiriakos N., Heidrich, Wolfgang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2020
• Subjects: Cameras; computational camera; deep neural network; Encoding; high-frame-rate imaging; Image reconstruction; Imaging; Network architecture; Numerical models; Optimization; Photography; Reconstruction algorithms; Training

Compressive imaging systems with spatial-temporal encoding can be used to capture and reconstruct fast-moving objects. The imaging quality highly depends on the choice of encoding masks and reconstruction methods. In this paper, we present a new network architecture to jointly design the encoding ma...