Cluster-level pseudo-labelling for source-free cross-domain facial expression recognition
This addresses the challenge of training models on sensitive face data without access to source datasets, which is incremental as it adapts existing SFUDA techniques to a specific domain.
The paper tackles the problem of domain shift and data sensitivity in facial expression recognition by proposing the first source-free unsupervised domain adaptation method for FER, which consistently outperforms existing SFUDA methods and matches UDA methods in performance.
Automatically understanding emotions from visual data is a fundamental task for human behaviour understanding. While models devised for Facial Expression Recognition (FER) have demonstrated excellent performances on many datasets, they often suffer from severe performance degradation when trained and tested on different datasets due to domain shift. In addition, as face images are considered highly sensitive data, the accessibility to large-scale datasets for model training is often denied. In this work, we tackle the above-mentioned problems by proposing the first Source-Free Unsupervised Domain Adaptation (SFUDA) method for FER. Our method exploits self-supervised pretraining to learn good feature representations from the target data and proposes a novel and robust cluster-level pseudo-labelling strategy that accounts for in-cluster statistics. We validate the effectiveness of our method in four adaptation setups, proving that it consistently outperforms existing SFUDA methods when applied to FER, and is on par with methods addressing FER in the UDA setting.