Self-Aware Adaptive Alignment: Enabling Accurate Perception for Intelligent Transportation Systems

2025-08-19

AI Summary

To address low cross-domain object detection accuracy in intelligent transportation systems, this paper proposes a self-perceiving and self-adaptive alignment framework. The method establishes a local-global collaborative feature alignment mechanism: (i) an attention-based channel re-weighting module dynamically recalibrates channel-wise importance of source-domain features; and (ii) a dual-granularity (instance-to-image-level) adaptive alignment strategy explicitly models target-domain-specific distributions. Integrated with a region proposal network and a cross-domain joint training paradigm, the framework enables end-to-end optimization without requiring target-domain annotations. Extensive experiments on major cross-domain detection benchmarks—including Cityscapes→Foggy Cityscapes and Sim10k→Cityscapes—demonstrate consistent superiority over state-of-the-art methods, achieving mAP gains of 3.2–5.7 percentage points. The proposed approach significantly enhances robust perception capability under complex traffic scenarios.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Adversarial Attacks & Robustness
• Application Domains: Transportation

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

Achieving top-notch performance in Intelligent Transportation detection is a critical research area. However, many challenges still need to be addressed when it comes to detecting in a cross-domain scenario. In this paper, we propose a Self-Aware Adaptive Alignment (SA3), by leveraging an efficient alignment mechanism and recognition strategy. Our proposed method employs a specified attention-based alignment module trained on source and target domain datasets to guide the image-level features alignment process, enabling the local-global adaptive alignment between the source domain and target domain. Features from both domains, whose channel importance is re-weighted, are fed into the region proposal network, which facilitates the acquisition of salient region features. Also, we introduce an instance-to-image level alignment module specific to the target domain to adaptively mitigate the domain gap. To evaluate the proposed method, extensive experiments have been conducted on popular cross-domain object detection benchmarks. Experimental results show that SA3 achieves superior results to the previous state-of-the-art methods.

Problem

• Addresses cross-domain object detection challenges in transportation
• Enhances feature alignment between source and target domains
• Improves perception accuracy for intelligent transportation systems

Innovation

• Self-Aware Adaptive Alignment mechanism
• Attention-based cross-domain feature alignment
• Instance-to-image level domain adaptation