DCDepth: Progressive Monocular Depth Estimation in Discrete Cosine Domain

TL;DR

DCDepth introduces a novel frequency domain approach for monocular depth estimation, progressively refining depth predictions by modeling local correlations in the discrete cosine domain, achieving state-of-the-art results.

Contribution

The paper presents a new framework that estimates depth in the frequency domain, enabling progressive refinement from global structure to local details.

Findings

Achieves state-of-the-art performance on NYU-Depth-V2, TOFDC, and KITTI datasets.

Demonstrates effective modeling of local depth correlations in the DCT domain.

Validates the superiority of frequency domain methods over traditional spatial domain approaches.

Abstract

In this paper, we introduce DCDepth, a novel framework for the long-standing monocular depth estimation task. Moving beyond conventional pixel-wise depth estimation in the spatial domain, our approach estimates the frequency coefficients of depth patches after transforming them into the discrete cosine domain. This unique formulation allows for the modeling of local depth correlations within each patch. Crucially, the frequency transformation segregates the depth information into various frequency components, with low-frequency components encapsulating the core scene structure and high-frequency components detailing the finer aspects. This decomposition forms the basis of our progressive strategy, which begins with the prediction of low-frequency components to establish a global scene context, followed by successive refinement of local details through the prediction of higher-frequency components. We conduct comprehensive experiments on NYU-Depth-V2, TOFDC, and KITTI datasets, and demonstrate the state-of-the-art performance of DCDepth. Code is available at https://github.com/w2kun/DCDepth.