Feature Augmentation based Test-Time Adaptation
It addresses data scarcity in TTA for practical environments, offering an incremental improvement by enhancing existing methods without architectural changes.
The paper tackles the problem of limited data in test-time adaptation (TTA) by proposing Feature Augmentation based Test-time Adaptation (FATA), which uses feature augmentation to fully utilize input data, resulting in validated superiority on benchmarks like ImageNet-C and Office-Home.
Test-time adaptation (TTA) allows a model to be adapted to an unseen domain without accessing the source data. Due to the nature of practical environments, TTA has a limited amount of data for adaptation. Recent TTA methods further restrict this by filtering input data for reliability, making the effective data size even smaller and limiting adaptation potential. To address this issue, We propose Feature Augmentation based Test-time Adaptation (FATA), a simple method that fully utilizes the limited amount of input data through feature augmentation. FATA employs Normalization Perturbation to augment features and adapts the model using the FATA loss, which makes the outputs of the augmented and original features similar. FATA is model-agnostic and can be seamlessly integrated into existing models without altering the model architecture. We demonstrate the effectiveness of FATA on various models and scenarios on ImageNet-C and Office-Home, validating its superiority in diverse real-world conditions.