Cross-Scale Cost Aggregation for Stereo Matching

TL;DR

This paper introduces a cross-scale cost aggregation framework inspired by human stereo processing, enhancing existing methods by enabling multi-scale interactions for improved dense stereo correspondence accuracy.

Contribution

A novel, unified optimization-based framework that integrates various cost aggregation methods through cross-scale regularization, inspired by biological vision.

Findings

Significant accuracy improvements on Middlebury, KITTI, and New Tsukuba datasets.

Framework effectively expands and enhances existing cost aggregation methods.

Demonstrates the importance of multi-scale interaction in stereo matching.

Abstract

Human beings process stereoscopic correspondence across multiple scales. However, this bio-inspiration is ignored by state-of-the-art cost aggregation methods for dense stereo correspondence. In this paper, a generic cross-scale cost aggregation framework is proposed to allow multi-scale interaction in cost aggregation. We firstly reformulate cost aggregation from a unified optimization perspective and show that different cost aggregation methods essentially differ in the choices of similarity kernels. Then, an inter-scale regularizer is introduced into optimization and solving this new optimization problem leads to the proposed framework. Since the regularization term is independent of the similarity kernel, various cost aggregation methods can be integrated into the proposed general framework. We show that the cross-scale framework is important as it effectively and efficiently expands state-of-the-art cost aggregation methods and leads to significant improvements, when evaluated on Middlebury, KITTI and New Tsukuba datasets.