Who's your daddy? A University of Central Florida (UCF) team programs computer to find out, in new research. A University of Central Florida research team has developed a facial recognition tool that promises to be useful in rapidly matching pictures of children with their biological parents and in potentially identifying photos of missing children as they age. A University of Central Florida research team has developed a facial recognition tool that promises to be useful in rapidly matching pictures of children with their biological parents and in potentially identifying photos of missing children as they age.
Researchers are going to present their findings at the annual CVPR: The IEEE Computer Vision and Pattern Recognition Conference which runs from June 23-27, 2014. You also may wish to check out the list of conferences and other presentations at the CVPR 2014 Webpage - PAMITC Conferences Page.
The work verifies that a computer is capable of matching pictures of parents and their children
The study will be presented at the nation's premier event for the science of computer vision - the IEEE Computer Vision and Pattern Recognition conference in Columbus, Ohio, which begins Monday, June 23, 2014. Graduate Student Afshin Dehfghan and a team from UCF's Center for Research in Computer Vision started the project with more than 10,000 online images of celebrities, politicians and their children.
Dehfghan says, according to the June 19, 2014 news release, Who's your daddy? UCF team programs computer to find out, "We wanted to see whether a machine could answer questions, such as 'Do children resemble their parents?' 'Do children resemble one parent more than another?' and 'What parts of the face are more genetically inspired?'"
Anthropologists have typically studied these questions. However Dehghan and his team are advancing a new wave of computational science that uses the power of a mechanical "mind" to evaluate data completely objectively – without the clutter of subjective human emotions and biases.
The power of a mechanical mind made to evaluate data completely objectively
The tool could be useful to law enforcement and families in locating missing children. "As this tool is developed I could see it being used to identify long-time missing children as they mature," says Ross Wolf, associate professor of criminal justice at the University of Central Florida (UCF).
Wolf said that facial recognition technology is already heavily used by law enforcement, but that it has not been developed to the point where it can identify the same characteristics in photos over time, something this technology could have the capability to do. Dehghan said he is planning to expand on the work in that area by studying how factors such as age and ethnicity affect the resemblance of facial features.
The project is the latest entry by UCF students into a scientific debate that has lived on for more than 60 years – can computers think?
Dehghan, who remembers being fascinated by a free-thinking robot in the movie The Terminator as he was growing up, says the project shows that computers can "deep think" by building on previous knowledge. While humans might look for something as prominent as actress Catherine Zeta Jones's smile in her offspring, the computer is able to focus on indicators people may not find as significant – such as the left eye, the chin and parts of the forehead. By designing an algorithm to focus on specific features, the research team converted the photos into a checkerboard of patches and extracted tiny snapshots of the most significant facial parts.
The computer compared all the photos feature by feature and sorted them by the most probable match. The team found that its program not only did a better job of matching features of parents and their kids than random chance, but it also outperformed existing software for identifying relatives through photos by 3 to 10 percent. The study affirmed that children resemble their parents, often in unseen ways, but that in the majority of cases (63 percent) sons resemble their fathers more than their mothers, and daughters are more likely to resemble their mothers (82 percent).
"Machines can learn through time," Dehghan explains, according to the news release. "When a computer goes through thousands of images it knows what it has seen and is able to tell you." Technology has grown to the point that computer scientists are able to revisit findings from traditional science fields and expand on them, said Mubarak Shah, according to the news release. Shah is one of the world's leading authorities in the emerging field of computer vision and director of the University of Central Florida (UCF) center.
Computer vision recognizes faces
Shah, who advised Dehghan and his colleagues on the project, says the technology could also be useful in areas such as homeland security, where authorities can use the technology to determine relationships between terrorists. The only issue with matching children and parents objectively, that consumers might think about is what happens when children are adopted and not genetically related to the parents in a situation where such a facial recognition tool might be used for homeland security to try to match closely related individuals.
Computer vision that recognizes faces in one application, but what are other possible uses you might think of where the technology can be used to objectively match data, genes, or the patterns in many types of data? One goal of the computer tool is matching pictures of parents and their children. What other uses could it have that you can list? One could be a dating service that matches you with lookalikes. See, "With Facial Recognition Partnership, Match.com Users Can Find Dates Based on Exes."
For example, that article notes that building on this concept, popular dating site Match.com has partnered with Los Angeles-based matchmaking service Three Day Rule that uses facial-recognition technology to find dates for users. But the researchers at the University of Central Florida in their unrelated research work had something else in mind with their computer vision research: For example homeland security applications. It goes to show you how wide the applications of facial recognition tools can be when the tool makes objective decisions. You may wish to savor the possibilities of how this technology can be used.
Other members of the research team are Enrique Ortiz, a graduate of the center, and Ruben Villegas, who completed his undergraduate work at UCF and is preparing to start a Ph.D. program at the University of Michigan. Enrique Ortiz and Afshin Dehfghan demonstrated the computer tool that matched images of Catherine Zeta-Jones with her children.