Deep Classification Network for Monocular Depth Estimation

TL;DR

This paper proposes a novel approach to monocular depth estimation by framing it as a pixel-level classification task, utilizing depth discretization and DeepLab v2, achieving state-of-the-art results on KITTI.

Contribution

It introduces a depth discretization method and applies DeepLab v2 to improve monocular depth estimation accuracy.

Findings

Achieved state-of-the-art results on KITTI dataset

Outperformed existing architectures by 8% margin

Validated the pixel classification perspective for depth estimation

Abstract

Monocular Depth Estimation is usually treated as a supervised and regression problem when it actually is very similar to semantic segmentation task since they both are fundamentally pixel-level classification tasks. We applied depth increments that increases with depth in discretizing depth values and then applied Deeplab v2 and the result was higher accuracy. We were able to achieve a state-of-the-art result on the KITTI dataset and outperformed existing architecture by an 8% margin.