<ul><li>Robust facial expression recognition in unconstrained, 'in-the-wild' environments remains challenging due to significant domain shifts between training and testing distributions.</li><li>Existing test-time adaptation (TTA) approaches often require manual selection of parameters to update, resulting in suboptimal adaptation and high computational costs.</li><li>A novel Fisher-driven selective adaptation framework is introduced, which dynamically identifies and updates only the most critical model parameters based on their importance as quantified by Fisher information.</li><li>The proposed approach achieves a significant improvement in F1 score over the base model, while adapting only a small subset of parameters, making test-time adaptation more efficient and practical for real-world affective computing applications.</li></ul>

FIESTA: Fisher Information-based Efficient Selective Test-time Adaptation

Discover more