Face recognition and analysis

This study evaluates how face-to-BMI Vision Transformer models perform across diverse human populations, finding that training on population-specific data improves accuracy, while global training data enhances performance in underrepresented groups.

Cheese3D detects and analyzes whole-face movements in mice with high precision, providing insights into physiological and neural states through 3D motion data.

This study improves face age prediction by using multiple-angle photos, showing that combining front and side views enhances accuracy compared to single-view models.

UniFacePoint-FM is a 3D facial foundation model that achieves state-of-the-art performance in gender classification, age regression, BMI prediction, and facial expression recognition.

This paper shows that training DNNs on face geometry and emotion improves generalization to new face tasks with fewer samples than standard models.

This paper develops a markerless 3D tracking system for speech movements with submillimeter accuracy, validated across adults and young children.

260,000+ primate images from 60+ genera with face boxes and 48/68-point landmarks; detectors and landmark models trained on them work across primates and approach human-trained baselines on human faces.

The paper uses face photos to predict mortality and shows that occupation affects age acceleration, with sports being the slowest aging occupation.

The paper develops an AI system for automated mouse facial image collection and sex determination, achieving 100% accuracy in classification.

The paper introduces the GestaltMatcher Database (GMDB), a global facial phenotypic reference for rare diseases, showing improved NGP performance with diverse ancestry data.

This paper compares three facial feature extraction methods for diagnosing genetic disorders, finding that VGGFace2 provides the best performance with 0.78 accuracy.

This paper develops a neural network classifier to diagnose genetic syndromes from facial images across different age groups, showing improved accuracy over clinical geneticists.

The paper introduces a convolutional neural network to detect defaced facial features in MRI brain scans, achieving high accuracy and sensitivity using data augmentation.

The paper proposes a feature transfer learning approach combining CNN and sparse approximation for face recognition with limited training data, achieving high accuracy on the LFW dataset.

This paper presents an automatic method to normalize and extract 3D facial meshes from head MRI scans, achieving high accuracy in orientation, alignment, and noise removal.