Research from Carnegie Mellon University Yields New Data on Machine Learning

摘要

"Classical face recognition techniques have been successful at operating under well-controlled nconditions; however, they have difficulty in robustly performing recognition in uncontrolled real-world nscenarios where variations in pose, illumination, and expression are encountered. In this paper, we npropose a new method for real-world unconstrained pose-invariant face recognition," scientists in nPittsburgh, Pennsylvania report.