Corpus ID: 235352538

Exploring Adversarial Learning for Deep Semi-Supervised Facial Action Unit Recognition

  title={Exploring Adversarial Learning for Deep Semi-Supervised Facial Action Unit Recognition},
  author={Shangfei Wang and Ya Chang and Guozhu Peng and Bowen Pan},
Current works formulate facial action unit (AU) recognition as a supervised learning problem, requiring fully AU-labeled facial images during training. It is challenging if not impossible to provide AU annotations for large numbers of facial images. Fortunately, AUs appear on all facial images, whether manually labeled or not, satisfy the underlying anatomic mechanisms and human behavioral habits. In this paper, we propose a deep semisupervised framework for facial action unit recognition from… Expand

Figures and Tables from this paper


Deep Facial Action Unit Recognition from Partially Labeled Data
This work proposes to learn label distribution from the ground-truth AU labels, and then train the AU classifiers from the large-scale facial images by maximizing the log likelihood of the mapping functions of AUs with regard to the learnt label distribution for all training data and minimizing the error between predicted AUs and ground- Truth AUs for labeled data simultaneously. Expand
Weakly Supervised Facial Action Unit Recognition Through Adversarial Training
This paper generates pseudo AU labels according to the probabilistic dependencies between expressions and AUs as well as correlations among AUs summarized from domain knowledge and proposes a weakly supervised AU recognition method via an adversarial process. Expand
Expression-assisted facial action unit recognition under incomplete AU annotation
A new learning method is introduced that trains an AU classifier using images with incomplete AU annotation but with complete expression labels to use expression labels as hidden knowledge to complement the missing AU labels. Expand
Exploring Domain Knowledge for Facial Expression-Assisted Action Unit Activation Recognition
A new AU activation recognition method is introduced that learns AU classifiers from domain knowledge, and requires only expression-annotated facial images, and can be extended to semi-supervised learning scenarios with partially AU-annotation facial images. Expand
Semantic Relationships Guided Representation Learning for Facial Action Unit Recognition
This paper investigates how to integrate the semantic relationship propagation between AUs in a deep neural network framework to enhance the feature representation of facial regions, and proposes an AU semantic relationship embedded representation learning (SRERL) framework. Expand
Facial action unit recognition under incomplete data based on multi-label learning with missing labels
Facial action unit recognition under incomplete data as a multi-label learning with missing labels (MLML) problem is formulated and the superiority of the proposed method in comparison with several state-of-the-art works is demonstrated. Expand
Confidence Preserving Machine for Facial Action Unit Detection
This paper proposes the confidence preserving machine (CPM), a novel two-stage learning framework that combines multiple classifiers following an “easy-to-hard” strategy, and proposes a quasi-semi-supervised (QSS) learning strategy to learn a person-specific classifier. Expand
Deep Region and Multi-label Learning for Facial Action Unit Detection
This paper proposes Deep Region and Multi-label Learning (DRML), a unified deep network that simultaneously addresses these two problems of facial Action Unit detection and region learning, allowing the two seemingly irrelevant problems to interact more directly. Expand
Classifier Learning with Prior Probabilities for Facial Action Unit Recognition
This work proposes a knowledge-driven method for jointly learning multiple AU classifiers without any AU annotation by leveraging prior probabilities on AUs, including expression-independent and expression-dependent AU probabilities. Expand
Deep Adaptive Attention for Joint Facial Action Unit Detection and Face Alignment
A novel end-to-end deep learning framework for joint AU detection and face alignment, which has not been explored before, in which multi-scale shared features are learned firstly, and high-level features of face alignment are fed into AU detection. Expand