Reinforcement Learning for Adaptive Video Compressive Sensing

TL;DR

This paper introduces a reinforcement learning approach to adaptively optimize the compression ratio in video snapshot compressive imaging, enabling real-time, low-cost, and scene-adaptive high-speed video capture.

Contribution

It proposes a novel RL-based framework for adaptive sensing in video SCI, including models for reconstruction and direct measurement-based object detection.

Findings

Achieves scene-adaptive compression ratio optimization

Enables real-time video SCI with low-cost implementation

Demonstrates improved object detection directly from measurements

Abstract

We apply reinforcement learning to video compressive sensing to adapt the compression ratio. Specifically, video snapshot compressive imaging (SCI), which captures high-speed video using a low-speed camera is considered in this work, in which multiple (B) video frames can be reconstructed from a snapshot measurement. One research gap in previous studies is how to adapt B in the video SCI system for different scenes. In this paper, we fill this gap utilizing reinforcement learning (RL). An RL model, as well as various convolutional neural networks for reconstruction, are learned to achieve adaptive sensing of video SCI systems. Furthermore, the performance of an object detection network using directly the video SCI measurements without reconstruction is also used to perform RL-based adaptive video compressive sensing. Our proposed adaptive SCI method can thus be implemented in low cost and real time. Our work takes the technology one step further towards real applications of video SCI.