Parallax to Align Them All: An OmniParallax Attention Mechanism for Distributed Multi-View Image Compression

TL;DR

This paper introduces OmniParallax Attention Mechanism and ParaHydra, a novel distributed multi-view image compression framework that significantly outperforms existing methods by effectively modeling inter-view correlations, especially as the number of views increases.

Contribution

The paper presents the first DMIC method that surpasses state-of-the-art MIC codecs using a new correlation modeling mechanism and adaptive information fusion.

Findings

Achieves up to 24.18% bitrate savings on WildTrack datasets.

Significantly improves encoding and decoding efficiency (up to 65x and 34x).

Outperforms existing DMIC methods as the number of views increases.

Abstract

Multi-view image compression (MIC) aims to achieve high compression efficiency by exploiting inter-image correlations, playing a crucial role in 3D applications. As a subfield of MIC, distributed multi-view image compression (DMIC) offers performance comparable to MIC while eliminating the need for inter-view information at the encoder side. However, existing methods in DMIC typically treat all images equally, overlooking the varying degrees of correlation between different views during decoding, which leads to suboptimal coding performance. To address this limitation, we propose a novel $\textbf{OmniParallax Attention Mechanism}$ (OPAM), which is a general mechanism for explicitly modeling correlations and aligned features between arbitrary pairs of information sources. Building upon OPAM, we propose a Parallax Multi Information Fusion Module (PMIFM) to adaptively integrate information from different sources. PMIFM is incorporated into both the joint decoder and the entropy model to construct our end-to-end DMIC framework, $\textbf{ParaHydra}$. Extensive experiments demonstrate that $\textbf{ParaHydra}$ is $\textbf{the first DMIC method}$ to significantly surpass state-of-the-art MIC codecs, while maintaining low computational overhead. Performance gains become more pronounced as the number of input views increases. Compared with LDMIC, $\textbf{ParaHydra}$ achieves bitrate savings of $\textbf{19.72%}$ on WildTrack(3) and up to $\textbf{24.18%}$ on WildTrack(6), while significantly improving coding efficiency (as much as $\textbf{65}\times$ in decoding and $\textbf{34}\times$ in encoding).