TTA-DAME: Test-Time Adaptation with Domain Augmentation and Model Ensemble for Dynamic Driving Conditions

2025-08-18

AI Summary

To address test-time domain shift induced by real-time variations in weather and illumination during dynamic driving, this paper proposes a test-time adaptation method that synergistically integrates domain augmentation with multi-detector ensemble learning. Domain augmentation simulates target-domain distributions using source-domain data; a domain discriminator and dedicated domain-specific detectors mitigate severe shifts (e.g., day–night transitions); and a weighted ensemble strategy combined with an improved non-maximum suppression (NMS) fuses outputs from multiple detectors. The method operates without target-domain labels and supports lightweight online updates. Evaluated on the SHIFT benchmark, it significantly enhances object detection robustness (mAP +3.2%), particularly under challenging conditions such as rain, fog, and nighttime. Its core contribution lies in the first unified modeling of domain augmentation, domain-aware detection, and ensemble-based test-time adaptation—enabling efficient, stable, and fully unsupervised adaptation in complex driving environments.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Test-time Adaptation (TTA) poses a challenge, requiring models to dynamically adapt and perform optimally on shifting target domains. This task is particularly emphasized in real-world driving scenes, where weather domain shifts occur frequently. To address such dynamic changes, our proposed method, TTA-DAME, leverages source domain data augmentation into target domains. Additionally, we introduce a domain discriminator and a specialized domain detector to mitigate drastic domain shifts, especially from daytime to nighttime conditions. To further improve adaptability, we train multiple detectors and consolidate their predictions through Non-Maximum Suppression (NMS). Our empirical validation demonstrates the effectiveness of our method, showing significant performance enhancements on the SHIFT Benchmark.

Problem

• Adapting models to dynamic driving conditions
• Handling weather and day-night domain shifts
• Improving robustness with ensemble and augmentation

Innovation

• Source domain data augmentation for target domains
• Domain discriminator and specialized domain detector
• Multiple detectors with NMS for consolidated predictions