A new 3D facial expression database has been developed to support research in facial behavior analysis. This database includes 2,500 3D facial expression models from 100 subjects, covering seven universal emotions with varying intensities. Each model is accompanied by 2D facial textures from two views, enabling both geometric and texture analysis. The database aims to advance research in affective computing and improve understanding of facial behavior and its 3D structure. It serves as a valuable resource for algorithm assessment, comparison, and evaluation. The database was created using a 3D face imaging system that captures facial expressions with high-resolution 3D data. Subjects performed seven expressions at four intensity levels, resulting in 25 models per subject. The data includes 3D models, textures, and additional descriptors for analysis. The database is structured for easy retrieval by gender, ethnicity, expression, and intensity. Validation experiments compared machine recognition, expert evaluation, and subject performance. The results showed high accuracy in expert recognition, with average rates of 94.1% for low intensity and 98.1% for highest intensity expressions. The database also includes 3D face pose information and feature points for segmentation and detection. The database has limitations, including the absence of dynamic expressions and limited expression types. Future work includes expanding the database with dynamic 3D expressions, more spontaneous expressions, and applications in medical and psychological research. The database is designed to be publicly accessible, allowing researchers to test and expand upon the data.A new 3D facial expression database has been developed to support research in facial behavior analysis. This database includes 2,500 3D facial expression models from 100 subjects, covering seven universal emotions with varying intensities. Each model is accompanied by 2D facial textures from two views, enabling both geometric and texture analysis. The database aims to advance research in affective computing and improve understanding of facial behavior and its 3D structure. It serves as a valuable resource for algorithm assessment, comparison, and evaluation. The database was created using a 3D face imaging system that captures facial expressions with high-resolution 3D data. Subjects performed seven expressions at four intensity levels, resulting in 25 models per subject. The data includes 3D models, textures, and additional descriptors for analysis. The database is structured for easy retrieval by gender, ethnicity, expression, and intensity. Validation experiments compared machine recognition, expert evaluation, and subject performance. The results showed high accuracy in expert recognition, with average rates of 94.1% for low intensity and 98.1% for highest intensity expressions. The database also includes 3D face pose information and feature points for segmentation and detection. The database has limitations, including the absence of dynamic expressions and limited expression types. Future work includes expanding the database with dynamic 3D expressions, more spontaneous expressions, and applications in medical and psychological research. The database is designed to be publicly accessible, allowing researchers to test and expand upon the data.