Bayesian dense inverse searching algorithm for real-time stereo matching in minimally invasive surgery

TL;DR

This paper introduces a real-time, CPU-efficient stereo matching algorithm using a Bayesian dense inverse searching approach, specifically designed for surgical images, achieving high accuracy and robustness against textureless surfaces and reflectance issues.

Contribution

The paper presents a novel Bayesian framework integrated with dense inverse searching for real-time stereo matching in surgical images, improving accuracy and robustness over existing methods.

Findings

Achieves 10 Hz processing on 640x480 images with a single CPU core.

Handles textureless surfaces and photometric inconsistencies effectively.

Outperforms baseline methods in accuracy and outlier reduction in surgical scenarios.

Abstract

This paper reports a CPU-level real-time stereo matching method for surgical images (10 Hz on 640 * 480 image with a single core of i5-9400). The proposed method is built on the fast ''dense inverse searching'' algorithm, which estimates the disparity of the stereo images. The overlapping image patches (arbitrary squared image segment) from the images at different scales are aligned based on the photometric consistency presumption. We propose a Bayesian framework to evaluate the probability of the optimized patch disparity at different scales. Moreover, we introduce a spatial Gaussian mixed probability distribution to address the pixel-wise probability within the patch. In-vivo and synthetic experiments show that our method can handle ambiguities resulted from the textureless surfaces and the photometric inconsistency caused by the Lambertian reflectance. Our Bayesian method correctly balances the probability of the patch for stereo images at different scales. Experiments indicate that the estimated depth has higher accuracy and fewer outliers than the baseline methods in the surgical scenario.