Back-Projection Pipeline

TL;DR

This paper introduces a multi-resolution residual network inspired by back-projection algorithms, designed to enhance images by learning to improve details across scales, demonstrating competitive results in super-resolution and raindrop removal.

Contribution

It presents a novel multi-resolution back-projection network using a data pipeline workflow, with features updated across scales via a system of ODEs, advancing image enhancement techniques.

Findings

Competitive performance on super-resolution tasks

Effective raindrop removal results

Strong ability to learn global and local features

Abstract

We propose a simple extension of residual networks that works simultaneously in multiple resolutions. Our network design is inspired by the iterative back-projection algorithm but seeks the more difficult task of learning how to enhance images. Compared to similar approaches, we propose a novel solution to make back-projections run in multiple resolutions by using a data pipeline workflow. Features are updated at multiple scales in each layer of the network. The update dynamic through these layers includes interactions between different resolutions in a way that is causal in scale, and it is represented by a system of ODEs, as opposed to a single ODE in the case of ResNets. The system can be used as a generic multi-resolution approach to enhance images. We test it on several challenging tasks with special focus on super-resolution and raindrop removal. Our results are competitive with state-of-the-arts and show a strong ability of our system to learn both global and local image features.