Variability as a route to understanding face recognition

This project represents a new way to look at the problem of human face recognition. Despite a large amount of research on this topic, we still do not understand the most fundamental aspect of face processing: how can we identify the people we see? This is a key problem in human perception, but it also has practical implications in forensic and security settings. This project has its roots in a simple observation: pictures of the same face can look very different indeed. In the standard approach to face recognition, this commonplace fact is treated as an inconvenience. Differences between pictures of the same person are regarded as 'noise', and either ignored or eliminated by systematically controlling the images used for research. This research programme takes exactly the converse approach. Instead of trying to control away this variability, it will be studied explicitly. Under this approach, the problem of face recognition is not how to 'tell people apart', but instead how to 'tell people together' - how to bring together superficially different images into a coherent representation. Early work suggests that a very important component of familiar face recognition is the ability to generalize over superficial image differences - differences which tend to fool unfamiliar viewers, as well as automatic computer-based systems. The current failure to address this variability may account for the slow progress in face identification - progress which has fallen behind the understanding of other aspects of face processing such as social perception. This research comprises three components. First, a systematic examination will be conducted of the physical differences between images of the same person. Applying statistical techniques to graphical data, the aim is to specify what aspects of face images vary commonly, and what aspects vary idiosyncratically to that person. Second, a series of behavioural experiments will examine the nature of our representations of familiar faces - the hypothesis is that this representation needs to incorporate variability. Third, a series of studies will address practical face recognition by human observers (e.g. for security purposes). Computer-based approaches will also be examined in this strand - as these systems remain very poor, despite the claims of vendors. This novel approach to face identification has the potential to make a significant contribution to an area which has progressed rather slowly in recent years.

The research planned here has direct relevance across a wide range of settings. Photo ID has become very common in the UK, in situations raging from the security-critical (e.g. proving one's identity at an airport) to the more prosaic (e.g. proving one's age in order to buy alcohol). Furthermore, the police and judicial system rely extensively on personal identification, for example when viewing crime-scene CCTV. Despite this reliance, it is simply an unreliably procedure: neither computers nor humans are good at matching unfamiliar people to their photos.

Part of the planned work directly addresses how to improve the use of photo-ID. As a regular speaker to police agencies, and as a teacher at the Scottish Police Training College, I will use these opportunities to describe the latest research. These forums are key, because one has direct access to people operational in the field. Existing contacts with the Criminal Cases Review Commission, and its Scottish counterpart will also be used to engage relevant professionals in the results of this work. The commissioners are particularly concerned with ID at the present, because matters of disputed identification are at the heart of several high profile legal cases. For this reason, both agencies have requested reports from my lab in the past.

My research group is involved in an on-going collaboration involving the University of New South Wales, and the Australian passport authority, and this is focused on the limits of facial identification (as well as potential improvements). This collaboration was originally funded by ESRC and the Australian Research Council, and has been very successful in passing research results to the Australian authorities. So far, the UK Border Agency has been less enthusiastic to engage with this research. However, in this project I undertake to develop contacts with UKBA, through the office of the Chief Scientific Adviser to the Home Office.

A further impact opportunity arises through the interest of engineers in the problem of face recognition. Current automated systems (for example at airports) work only very poorly. The theory driving this proposal is that the typical engineering approach is unlikely to produce robust levels of recognition, and that advances in understanding human perceptual processes can improve automated systems. For this reason, I have requested support to disseminate research on human perceptual face processing to engineering audiences (in both private and public sectors).

Throughout the period, I will continue to engage a broader audience with this research. I am regularly asked to speak at public engagement events, and to the media, and so there will be many opportunities to do this.