A taxonomy and evaluation of dense two-frame stereo correspondence algorithms

TLDR

Stereo matching is a highly active research area, yet performance characterization of existing algorithms remains limited. This paper introduces a taxonomy of dense, two‑frame stereo methods to evaluate design choices in individual algorithms. The authors built a flexible C++ platform and curated datasets, using the taxonomy to compare and benchmark various stereo methods. Experiments reveal performance differences across variants, and the authors released new multiframe datasets, code, and the evaluation framework online.

Abstract

Stereo matching is one of the most active research areas in computer vision. While a large number of algorithms for stereo correspondence have been developed, relatively little work has been done on characterizing their performance. In this paper, we present a taxonomy of dense, two-frame stereo methods designed to assess the different components and design decisions made in individual stereo algorithms. Using this taxonomy, we compare existing stereo methods and present experiments evaluating the performance of many different variants. In order to establish a common software platform and a collection of data sets for easy evaluation, we have designed a stand-alone, flexible C++ implementation that enables the evaluation of individual components and that can be easily extended to include new algorithms. We have also produced several new multiframe stereo data sets with ground truth, and are making both the code and data sets available on the Web.