Evaluation of CNN-based Single-Image Depth Estimation Methods

TL;DR

This paper introduces new evaluation metrics and a high-quality dataset to better assess the performance of CNN-based single-image depth estimation methods, focusing on edge preservation, planar regions, and accuracy.

Contribution

It proposes novel quality criteria and provides a new high-resolution RGB-D dataset for comprehensive evaluation of depth estimation techniques.

Findings

The proposed metrics effectively analyze depth map quality.

The dataset enables detailed comparison of state-of-the-art methods.

Experimental results validate the evaluation protocol.

Abstract

While an increasing interest in deep models for single-image depth estimation methods can be observed, established schemes for their evaluation are still limited. We propose a set of novel quality criteria, allowing for a more detailed analysis by focusing on specific characteristics of depth maps. In particular, we address the preservation of edges and planar regions, depth consistency, and absolute distance accuracy. In order to employ these metrics to evaluate and compare state-of-the-art single-image depth estimation approaches, we provide a new high-quality RGB-D dataset. We used a DSLR camera together with a laser scanner to acquire high-resolution images and highly accurate depth maps. Experimental results show the validity of our proposed evaluation protocol.