Project Team
Students
Bohan Dong
Computer Science
Penn State Abington
Faculty Mentors
Vinayak Elangovan
Penn State Abington
Computer Science
Banyay Gregory
Penn State University Park
Fluid Dynamics and Acoustics
Project
Project Video
Project Abstract
After society has entered the Era of science and technology, face recognition technology has been more and more widely used. Face recognition technology has been indispensable from daily unlocking mobile phones to important occasions for rapid identification of people. At the time of storage face data, 3D model file has more advantages than 2D image file, such as each person only need one file can store all the facial information, and 3D models when human faces than identify more quickly and efficiently, so I study integrating 2D image and converted into 3D model. In this research, we explored various approaches for generating a 3D model of a face from 2D image/s. In particular, we focused more on two important techniques namely, Volume reconstruction and Feature extraction techniques. These techniques were implemented, and the results are compared. According to my research results, Volume reconstruction has more detail and accuracy, but on the contrary, this technique requires more image quantity to make the 3D model have more detail, and the output 3D model with fewer images is of poor quality. And this technique is very time consuming. For Feature extraction, this technique only needs one image to output a 3D model, and the speed is fast. However, it needs many times of training on the facial database to achieve certain accuracy and details, and generally the details are poor.
Evaluate this Project
Use this form link to provide feedback to the presenters, and add your project evaluation for award(s) consideration.