The media industry is currently facing a major challenge in that digitisation, the burgeoning number of television channels, and the growth of the Internet are generating an ever-increasing need for appropriate multimedia content. Although extensive audio-visual archives are already in existence, rapid access to these media archives as well as flexible digital processing of the data are still limited, especially with regard to video material. Video archiving systems are predominantly comprised of purely textual databases, and accessing archived material in this way is extremely time-consuming.

In order to analyse and manage large collections of digital video information in multimedia information systems and to support content-based navigation, Songyang Lao's research focuses on video content analysis and understanding, video summarisation and visualisation, content-based indexing, search and retrieval of video. Recently, Dr. Lao has been developing tools to (semi) automatically analyse and index digital video information (particularly in digital video of news and football matches) and allow content-based operations such as browsing, searching and summarisation.

Other research areas in which Dr. Lao is interested include learning and relevance feedback in image/video retrieval, database architectures for image/video retrieval, multimedia information search and browsing on the web.