Distortion-Tolerant Monocular Depth Estimation On Omnidirectional Images Using Dual-cubemap

TL;DR

This paper introduces a novel distortion-tolerant depth estimation method for omnidirectional images using a dual-cubemap approach, improving accuracy and boundary continuity over existing techniques.

Contribution

It proposes a dual-cubemap model with boundary revision to effectively handle distortion and boundary discontinuities in omnidirectional depth estimation.

Findings

Outperforms state-of-the-art methods in accuracy

Reduces boundary discontinuities in depth maps

Enhances visual continuity of omnidirectional depths

Abstract

Estimating the depth of omnidirectional images is more challenging than that of normal field-of-view (NFoV) images because the varying distortion can significantly twist an object's shape. The existing methods suffer from troublesome distortion while estimating the depth of omnidirectional images, leading to inferior performance. To reduce the negative impact of the distortion influence, we propose a distortion-tolerant omnidirectional depth estimation algorithm using a dual-cubemap. It comprises two modules: Dual-Cubemap Depth Estimation (DCDE) module and Boundary Revision (BR) module. In DCDE module, we present a rotation-based dual-cubemap model to estimate the accurate NFoV depth, reducing the distortion at the cost of boundary discontinuity on omnidirectional depths. Then a boundary revision module is designed to smooth the discontinuous boundaries, which contributes to the precise and visually continuous omnidirectional depths. Extensive experiments demonstrate the superiority of our method over other state-of-the-art solutions.