Cross-MPI: Cross-scale Stereo for Image Super-Resolution using Multiplane Images

TL;DR

Cross-MPI introduces a novel end-to-end network leveraging scene structure-aware attention and multiscale guided upsampling to significantly improve reference-based super-resolution in multiscale imaging systems with large resolution gaps.

Contribution

The paper proposes Cross-MPI, a new RefSR framework that effectively utilizes scene structure and multiscale guidance for high-fidelity super-resolution across large resolution differences.

Findings

Outperforms existing RefSR methods on synthesized and optical zoom data.

Achieves robust and accurate detail transmission in multiscale camera systems.

Demonstrates superior performance with large-scale resolution gaps.

Abstract

Various combinations of cameras enrich computational photography, among which reference-based superresolution (RefSR) plays a critical role in multiscale imaging systems. However, existing RefSR approaches fail to accomplish high-fidelity super-resolution under a large resolution gap, e.g., 8x upscaling, due to the lower consideration of the underlying scene structure. In this paper, we aim to solve the RefSR problem in actual multiscale camera systems inspired by multiplane image (MPI) representation. Specifically, we propose Cross-MPI, an end-to-end RefSR network composed of a novel plane-aware attention-based MPI mechanism, a multiscale guided upsampling module as well as a super-resolution (SR) synthesis and fusion module. Instead of using a direct and exhaustive matching between the cross-scale stereo, the proposed plane-aware attention mechanism fully utilizes the concealed scene structure for efficient attention-based correspondence searching. Further combined with a gentle coarse-to-fine guided upsampling strategy, the proposed Cross-MPI can achieve a robust and accurate detail transmission. Experimental results on both digitally synthesized and optical zoom cross-scale data show that the Cross-MPI framework can achieve superior performance against the existing RefSR methods and is a real fit for actual multiscale camera systems even with large-scale differences.