ALDI-ray: Adapting the ALDI Framework for Security X-ray Object Detection

TL;DR

This paper introduces ALDI++, an advanced domain adaptation framework that significantly improves security X-ray object detection across different environments, leveraging self-distillation, feature alignment, and transformer architectures.

Contribution

We adapt and enhance the ALDI framework with new training strategies and transformer-based models, achieving state-of-the-art results in cross-domain security X-ray detection.

Findings

ALDI++ outperforms existing domain adaptation methods on the EDS dataset.

Transformer-based ViTDet backbone yields the highest mean average precision.

Consistent category-wise detection improvements demonstrate robustness.

Abstract

Domain adaptation in object detection is critical for real-world applications where distribution shifts degrade model performance. Security X-ray imaging presents a unique challenge due to variations in scanning devices and environmental conditions, leading to significant domain discrepancies. To address this, we apply ALDI++, a domain adaptation framework that integrates self-distillation, feature alignment, and enhanced training strategies to mitigate domain shift effectively in this area. We conduct extensive experiments on the EDS dataset, demonstrating that ALDI++ surpasses the state-of-the-art (SOTA) domain adaptation methods across multiple adaptation scenarios. In particular, ALDI++ with a Vision Transformer for Detection (ViTDet) backbone achieves the highest mean average precision (mAP), confirming the effectiveness of transformer-based architectures for cross-domain object detection. Additionally, our category-wise analysis highlights consistent improvements in detection accuracy, reinforcing the robustness of the model across diverse object classes. Our findings establish ALDI++ as an efficient solution for domain-adaptive object detection, setting a new benchmark for performance stability and cross-domain generalization in security X-ray imagery.