Below is an abstract from a paper published in 2005 by Kui Jia and Shaogang Gong both at the Queen Mary University, London.
These papers show, along with other research by the Queen Mary University, some of which is MoD funded, how much technology is now available to CCTV systems. The simple “pan and tilt” options are a very “last millennium” way of thinking for CCTV.
In this paper, we present a novel learning-based algorithm to super-resolve multiple partially occluded CCTV low-resolution face images. By integrating hierarchical patchwise alignment and inter-frame constraints into a Bayesian framework, we can probabilistically align multiple inputimages at different resolutions and recursively infer the high-resolution face image. We address the problem of fusing partial imagery information through multiple frames and discuss the new algorithm’s effectiveness when encountering occluded low-resolution face images. We show promising results compared to that of existing face hallucination methods.
Full paper is available here – cctv-paper-resolution