Analysis of different disparity estimation techniques on aerial stereo image datasets

TL;DR

This paper compares traditional, optimization-based, and learning-based disparity estimation techniques on aerial stereo datasets, providing insights into their performance and benchmarking in this specific domain.

Contribution

It offers a comprehensive analysis and benchmarking of various disparity estimation methods on aerial images, including implementation details and performance evaluation.

Findings

Traditional methods perform well but lack aerial dataset benchmarking.

Learning-based methods show promising results with pre-trained models.

Different cost functions significantly impact depth estimation accuracy.

Abstract

With the advent of aerial image datasets, dense stereo matching has gained tremendous progress. This work analyses dense stereo correspondence analysis on aerial images using different techniques. Traditional methods, optimization based methods and learning based methods have been implemented and compared here for aerial images. For traditional methods, we implemented the architecture of Stereo SGBM while using different cost functions to get an understanding of their performance on aerial datasets. Analysis of most of the methods in standard datasets has shown good performance, however in case of aerial dataset, not much benchmarking is available. Visual qualitative and quantitative analysis has been carried out for two stereo aerial datasets in order to compare different cost functions and techniques for the purpose of depth estimation from stereo images. Using existing pre-trained models, recent learning based architectures have also been tested on stereo pairs along with different cost functions in SGBM. The outputs and given ground truth are compared using MSE, SSIM and other error metrics.