Meta-UDA: Unsupervised Domain Adaptive Thermal Object Detection using Meta-Learning

TL;DR

This paper introduces a meta-learning based framework to improve unsupervised domain adaptation for thermal object detection, leveraging labeled visible data to enhance performance in thermal imagery, especially under challenging lighting conditions.

Contribution

It proposes an online meta-learning approach to optimize detector initialization for better domain adaptation without designing new UDA strategies.

Findings

Achieves state-of-the-art results on KAIST and DSIAC datasets.

Outperforms baseline methods in thermal object detection accuracy.

Demonstrates effectiveness of online meta-learning in reducing computational complexity.

Abstract

Object detectors trained on large-scale RGB datasets are being extensively employed in real-world applications. However, these RGB-trained models suffer a performance drop under adverse illumination and lighting conditions. Infrared (IR) cameras are robust under such conditions and can be helpful in real-world applications. Though thermal cameras are widely used for military applications and increasingly for commercial applications, there is a lack of robust algorithms to robustly exploit the thermal imagery due to the limited availability of labeled thermal data. In this work, we aim to enhance the object detection performance in the thermal domain by leveraging the labeled visible domain data in an Unsupervised Domain Adaptation (UDA) setting. We propose an algorithm agnostic meta-learning framework to improve existing UDA methods instead of proposing a new UDA strategy. We achieve this by meta-learning the initial condition of the detector, which facilitates the adaptation process with fine updates without overfitting or getting stuck at local optima. However, meta-learning the initial condition for the detection scenario is computationally heavy due to long and intractable computation graphs. Therefore, we propose an online meta-learning paradigm which performs online updates resulting in a short and tractable computation graph. To this end, we demonstrate the superiority of our method over many baselines in the UDA setting, producing a state-of-the-art thermal detector for the KAIST and DSIAC datasets.