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

TL;DR

TTA-DAME enhances test-time adaptation in dynamic driving scenarios by using domain augmentation, ensemble detectors, and domain discrimination to handle weather-related domain shifts.

Contribution

It introduces a novel TTA method combining domain augmentation, multiple detectors, and domain discrimination to improve adaptation in shifting driving conditions.

Findings

Significant performance improvements on the SHIFT Benchmark.

Effective mitigation of drastic domain shifts from day to night.

Enhanced adaptability in real-world driving scenes.

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.