The Role of Cyclopean-Eye in Stereo Vision

TL;DR

This paper explores the geometric principles of stereo vision, emphasizing the Cyclopean Eye model, and combines geometric constraints with deep learning features to improve 3D depth reconstruction.

Contribution

It introduces novel geometric constraints based on the Cyclopean Eye model and integrates deep learning features and attention mechanisms for enhanced stereo vision analysis.

Findings

Geometric constraints improve depth accuracy

Deep learning features enhance feature matching

Attention mechanisms recover meaningful 3D surfaces

Abstract

This work investigates the geometric foundations of modern stereo vision systems, with a focus on how 3D structure and human-inspired perception contribute to accurate depth reconstruction. We revisit the Cyclopean Eye model and propose novel geometric constraints that account for occlusions and depth discontinuities. Our analysis includes the evaluation of stereo feature matching quality derived from deep learning models, as well as the role of attention mechanisms in recovering meaningful 3D surfaces. Through both theoretical insights and empirical studies on real datasets, we demonstrate that combining strong geometric priors with learned features provides internal abstractions for understanding stereo vision systems.