UK facial recognition project to identify hidden faces

Collaboration

The FACER2VM project has brought together experts from the University of Surrey, Imperial College London and the University of Stirling, as well as collaborators from a range of companies and public sector bodies.

Companies include IBM, Digital Barriers and Cognitec Systems, while public sector actors include the Home Office (an official project partner) and the MPS.

According to information listed on UK Research and Innovation (UKRI), the project coordinators expect their research to have a substantial impact.

“The societal impact is anticipated to be multifaceted. Unconstrained face biometrics capability will significantly contribute to the government’s security agenda in the framework of smart cities and national security. It can effectively facilitate access to public services,” it said.

It added that its impact would be maximised by the Home Office Centre for Applied Science and Technology (Cast), as well as its user group.  

However, while the tech could have a number of potential uses, the EPSRC has classified the project’s work as being for the “aerospace, defence and marine” industrial sector, according to the research grant application form.

Potential use against protestors

In response to previous questions from Computer Weekly, the MPS neither confirmed nor denied that its LFR technology would be used in a protest setting.

However, the technology has already been used against protestors in the UK by South Wales Police, and officers across Britain now routinely film gatherings and demonstrations.

The Network for Police Monitoring (Netpol), which monitors and resists policing that is excessive, discriminatory or threatens civil liberties, told Computer Weekly at the time that individuals who seek to conceal their identities to evade LFR on privacy grounds could unwittingly garner more attention from police who may assume they are “troublemakers”.

It could also dissuade them from participating in political action all together, it added.

Due to FACER2VM’s focus on “image degradation phenomena”, the research will actively seek to produce LFR technology that combats “noise, blur and occlusion”, which, if successful, would allow the systems to identify people with covered faces.

According collaboration information on UKRI, the Home Office is interested in the surveillance capabilities of FACER2VM, and contributed to the project through Cast, providing a “proprietary database of videos representative of the types of surveillance scenarios of interest to Home Office”.

For some, the involvement of Jiangnan University has sparked fears the project could enhance the Chinese government’s ability to identify masked protesters in Hong Kong, as well as Uighur Muslims in Xinjiang.

Commenting directly on the FACEER2VM project, Netpol coordinator Kevin Blowe said there are numerous reasons why protestors may choose to cover their faces, ranging from resisting authoritarian governments that want to build up a profile of opposition to its policies, to people worrying about their employment or education opportunities.

“These concerns are far from unique to the conduct of regimes like China: all have been features of policing in Britain over the past decade, from spying on protest movements to the use of Prevent to target Muslim communities,” he said. “That’s all the more important to remember at a moment, when the government in Britain is planning the most sweeping new emergency powers in 70 years.”

“When it comes to speaking out, anonymity may represent the main deciding factor for many about whether they are able to exercise their rights or not,” said Blowe. “What is apparently missing, however, is enough academics involved in this kind of research asking what the ethics of our work are. Should we be doing this?”

The FACER2VM project will run until 30 September 2021.