Intelligent CCTV for Mass Transport Security: Challenges and Opportunities for Video and Face Processing

CCTV surveillance systems have long been promoted as being e ffective in improving public safety. However due to the amount of cameras installed, many sites ha ve abandoned expensive human monitoring and only record video for forensic purposes. One of the sough t-after capabilities of an automated surveillance system is “face in the crowd” recognition, in public sp aces such as mass transit centres. Apart from accuracy and robustness to nuisance factors such as pose var iations, in such surveillance situations the other important factors are scalability and fast performance. We evaluate recent approaches to the recognition of faces at large pose angles from a gallery of frontal images an d propose novel adaptations as well as modifications. We compare and contrast the accuracy, robustnes s and speed of an Active Appearance Model (AAM) based method (where realistic frontal faces are synth esized from non-frontal probe faces) against bag-of-features methods. We show a novel approach where the performance of the AAM based technique is increased by side-stepping the image synthesis step, als o resulting in a considerable speedup. Additionally, we adapt a histogram-based bag-of-features techniqu e to face classification and contrast its properties to a previously proposed direct bag-of-features method. We further show that the two bag-of-features approaches can be considerably sped up, without a loss in class ific tion accuracy, via an approximation of the exponential function. Experiments on the FERET and PIE data bases suggest that the bag-of-features techniques generally attain better performance, with significa ntly lower computational loads. The histogrambased bag-of-features technique is capable of achieving an average recognition accuracy of 89% for pose angles of around 25 degrees. Finally, we provide a discussio n on implementation as well as legal challenges surrounding research on automated surveillance.