# Refine and Distill: Exploiting Cycle-Inconsistency and Knowledge Distillation for Unsupervised Monocular Depth Estimation

## TL;DR

This paper introduces a self-supervised monocular depth estimation model that leverages cycle-inconsistency and knowledge distillation, outperforming existing unsupervised methods on the KITTI benchmark.

## Contribution

The novel framework combines cycle-inconsistency refinement with knowledge distillation to improve unsupervised depth estimation accuracy.

## Key findings

- Outperforms state-of-the-art unsupervised methods on KITTI
- Effective use of cycle-inconsistency for depth refinement
- Knowledge distillation enhances model performance

## Abstract

Nowadays, the majority of state of the art monocular depth estimation techniques are based on supervised deep learning models. However, collecting RGB images with associated depth maps is a very time consuming procedure. Therefore, recent works have proposed deep architectures for addressing the monocular depth prediction task as a reconstruction problem, thus avoiding the need of collecting ground-truth depth. Following these works, we propose a novel self-supervised deep model for estimating depth maps. Our framework exploits two main strategies: refinement via cycle-inconsistency and distillation. Specifically, first a \emph{student} network is trained to predict a disparity map such as to recover from a frame in a camera view the associated image in the opposite view. Then, a backward cycle network is applied to the generated image to re-synthesize back the input image, estimating the opposite disparity. A third network exploits the inconsistency between the original and the reconstructed input frame in order to output a refined depth map. Finally, knowledge distillation is exploited, such as to transfer information from the refinement network to the student. Our extensive experimental evaluation demonstrate the effectiveness of the proposed framework which outperforms state of the art unsupervised methods on the KITTI benchmark.