Since many commercially available 3-D sensors suitable for face image capture employ passive or texture-assisted stereo imaging or structured illumination with a moving light stripe, these techniques require a stationary subject, so the authors of this paper describe an initial design and evaluation of a fixed-stripe moving object 3-D scanner designed for human faces.
The authors’ method of acquisition requires the subject to walk through a static light screen generated by two laser line projectors. Triangulation and tracking applied to the video sequences captured during subject motion yield a 3-D image of the subject's face from multiple images. To demonstrate the accuracy of the authors’ initial design, a small-scale facial recognition experiment was executed. In an experiment involving 81 subjects with four images per subject on the average, the authors used two gallery images per subject, and achieved 89.6% rank-one recognition using an iterative closest point (ICP)-based matching method, demonstrating the feasibility of the technique. (Publisher abstract provided)
Similar Publications
- Scaling up Schoolwide Positive Behavioral Interventions and Supports (SWPBIS) Framework in Rural Settings Through the Idaho Rural Implementation Model
- Machine Learning and the Prevention of Mass Shooting in the United States
- A Self-assessment Tool for Helping Identify Police Burnout Among Investigators of Child Sexual Abuse Material