Thermal Infrared Colorization via Conditional Generative Adversarial Network
This work addresses a domain-specific challenge in computer vision for applications like surveillance or medical imaging, but it is incremental as it builds on existing GAN-based methods.
The paper tackles the problem of transforming thermal infrared images into realistic RGB images by proposing a deep learning method with a coarse-to-fine generator and a composite loss function, resulting in significant outperformance over existing approaches in quantitative and qualitative experiments.
Transforming a thermal infrared image into a realistic RGB image is a challenging task. In this paper we propose a deep learning method to bridge this gap. We propose learning the transformation mapping using a coarse-to-fine generator that preserves the details. Since the standard mean squared loss cannot penalize the distance between colorized and ground truth images well, we propose a composite loss function that combines content, adversarial, perceptual and total variation losses. The content loss is used to recover global image information while the latter three losses are used to synthesize local realistic textures. Quantitative and qualitative experiments demonstrate that our approach significantly outperforms existing approaches.