The Power of Properties: Uncovering the Influential Factors in Emotion Classification

arXiv:2404.07867v1 Announce Type: new
Abstract: Facial expression-based human emotion recognition is a critical research area in psychology and medicine. State-of-the-art classification performance is only reached by end-to-end trained neural networks. Nevertheless, such black-box models lack transparency in their decision-making processes, prompting efforts to ascertain the rules that underlie classifiers' decisions. Analyzing single inputs alone fails to expose systematic learned biases. These biases can be characterized as facial properties summarizing abstract information like age or medical conditions. Therefore, understanding a model's …

abstract art arxiv box classification classifiers cs.cv decision emotion facial expression human making medicine networks neural networks performance power processes prompting psychology recognition research rules state transparency type