ReconNet: Non-Iterative Reconstruction of Images from Compressively Sensed Random Measurements
This work addresses the need for faster and more robust image reconstruction in compressive sensing applications, such as real-time visual tracking, though it is incremental as it builds on existing CNN and denoising methods.
The paper tackles the problem of reconstructing images from compressively sensed random measurements by proposing ReconNet, a non-iterative convolutional neural network that outputs intermediate reconstructions, which are then denoised, resulting in significant improvements in PSNR and time complexity over state-of-the-art iterative algorithms at various measurement rates, with robustness to sensor noise and recovery of visually better images at rates as low as 0.04.
The goal of this paper is to present a non-iterative and more importantly an extremely fast algorithm to reconstruct images from compressively sensed (CS) random measurements. To this end, we propose a novel convolutional neural network (CNN) architecture which takes in CS measurements of an image as input and outputs an intermediate reconstruction. We call this network, ReconNet. The intermediate reconstruction is fed into an off-the-shelf denoiser to obtain the final reconstructed image. On a standard dataset of images we show significant improvements in reconstruction results (both in terms of PSNR and time complexity) over state-of-the-art iterative CS reconstruction algorithms at various measurement rates. Further, through qualitative experiments on real data collected using our block single pixel camera (SPC), we show that our network is highly robust to sensor noise and can recover visually better quality images than competitive algorithms at extremely low sensing rates of 0.1 and 0.04. To demonstrate that our algorithm can recover semantically informative images even at a low measurement rate of 0.01, we present a very robust proof of concept real-time visual tracking application.