Incremental Boosting Convolutional Neural Network for Facial Action Unit Recognition
This work addresses the challenge of limited training data for facial AU recognition, which is important for applications in affective computing and human-computer interaction, but it is incremental as it builds on existing CNN and boosting techniques.
The paper tackled the problem of overfitting and poor generalization in facial action unit (AU) recognition from spontaneous expressions by proposing an Incremental Boosting CNN (IB-CNN), which integrates boosting via an incremental layer and a novel loss function, resulting in significant improvements over traditional and state-of-the-art CNN-based methods, especially for low-frequency AUs.
Recognizing facial action units (AUs) from spontaneous facial expressions is still a challenging problem. Most recently, CNNs have shown promise on facial AU recognition. However, the learned CNNs are often overfitted and do not generalize well to unseen subjects due to limited AU-coded training images. We proposed a novel Incremental Boosting CNN (IB-CNN) to integrate boosting into the CNN via an incremental boosting layer that selects discriminative neurons from the lower layer and is incrementally updated on successive mini-batches. In addition, a novel loss function that accounts for errors from both the incremental boosted classifier and individual weak classifiers was proposed to fine-tune the IB-CNN. Experimental results on four benchmark AU databases have demonstrated that the IB-CNN yields significant improvement over the traditional CNN and the boosting CNN without incremental learning, as well as outperforming the state-of-the-art CNN-based methods in AU recognition. The improvement is more impressive for the AUs that have the lowest frequencies in the databases.