Maximizing Self-supervision from Thermal Image for Effective Self-supervised Learning of Depth and Ego-motion

TL;DR

This paper enhances self-supervised learning of depth and ego-motion from thermal images by proposing a mapping method that improves image quality and detail, leading to better results without needing extra RGB data.

Contribution

It introduces a novel thermal image mapping technique that boosts information content and maintains temporal consistency, improving depth and pose estimation accuracy.

Findings

Outperforms previous state-of-the-art networks in depth and pose estimation.

Effectively increases thermal image information such as structure and contrast.

Achieves better results without additional RGB guidance.

Abstract

Recently, self-supervised learning of depth and ego-motion from thermal images shows strong robustness and reliability under challenging scenarios. However, the inherent thermal image properties such as weak contrast, blurry edges, and noise hinder to generate effective self-supervision from thermal images. Therefore, most research relies on additional self-supervision sources such as well-lit RGB images, generative models, and Lidar information. In this paper, we conduct an in-depth analysis of thermal image characteristics that degenerates self-supervision from thermal images. Based on the analysis, we propose an effective thermal image mapping method that significantly increases image information, such as overall structure, contrast, and details, while preserving temporal consistency. The proposed method shows outperformed depth and pose results than previous state-of-the-art networks without leveraging additional RGB guidance.