Thermal Infrared Image Inpainting via Edge-Aware Guidance

TL;DR

This paper introduces TIR-Fill, a deep learning model specifically designed for thermal infrared image inpainting, effectively reconstructing missing regions by leveraging edge guidance and a refinement network, outperforming existing methods.

Contribution

The paper presents a novel thermal infrared image inpainting method that incorporates edge-aware guidance and a gated convolution refinement network, tailored for TIR images.

Findings

Outperforms state-of-the-art inpainting methods on FLIR dataset

Effectively reconstructs missing regions in TIR images

Enhances edge awareness and image consistency

Abstract

Image inpainting has achieved fundamental advances with deep learning. However, almost all existing inpainting methods aim to process natural images, while few target Thermal Infrared (TIR) images, which have widespread applications. When applied to TIR images, conventional inpainting methods usually generate distorted or blurry content. In this paper, we propose a novel task -- Thermal Infrared Image Inpainting, which aims to reconstruct missing regions of TIR images. Crucially, we propose a novel deep-learning-based model TIR-Fill. We adopt the edge generator to complete the canny edges of broken TIR images. The completed edges are projected to the normalization weights and biases to enhance edge awareness of the model. In addition, a refinement network based on gated convolution is employed to improve TIR image consistency. The experiments demonstrate that our method outperforms state-of-the-art image inpainting approaches on FLIR thermal dataset.