The Curious Face Off

Curious Cases

BBC

Science

4.8 • 4.1K Ratings

🗓️ 5 January 2018

⏱️ 26 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

"Are machines better than humans at identifying faces?" asks the excellently named Carl Vandal.

Today's Face Off leads our intrepid detectives to investigate why we see Jesus on toast, Hitler in houses and Kate Middleton on a jelly bean.

Face perception psychologist Rob Jenkins from the University of York explains why we're so good at spotting familiar faces, like celebrities. Plus, Franziska Knolle from the University of Cambridge discusses her face recognition study involving Barack Obama and a group of highly-trained sheep.

But are we outwitted by artificial intelligence when it comes to face ID? BBC technology correspondent Rory Cellan-Jones gives us the low-down on the pros and cons of current technology.

Presenters: Hannah Fry, Adam Rutherford Producer: Michelle Martin.

Transcript

Click on a timestamp to play from that location

0:00.0	I'm Dr Adam Rutherford. And I'm Dr Hannah Fry. And you are going to send us your everyday
0:10.8	mysteries. And we are going to investigate them using the power of science. Science.
0:16.5	I like it. Hello and Happy New Year everyone. This podcast of the Curious Cases was first
0:26.3	released on the 5th of January 2018, hence the Happy New Year in case you've downloaded
0:31.4	it at a different time and are confused. Although now your explanation for being confused
0:36.3	is taking longer than the wishing of Happy New Year in the first place. But I'm just very
0:41.4	polite. Clearly. Anyway, in this episode we feature a question from Electro Khan, as I think
0:48.9	he's now uniformly named. Yes, you may remember him from our static shocks episodes.
0:53.6	That's right. Karl Vandal, he came onto the pods and we interviewed him because he believed
0:58.6	he was generating so much static electricity in his normal everyday life that he would give
1:03.6	his wife and kids and whole family shocks during just normal day to day living. I seem to
1:09.3	remember him claiming that they have banned him from touching them. Which sounds much worse
1:13.5	than it actually is. So let us move on. This question from Karl Vandal is all about our
1:18.5	faces. It is. And specifically where their humans can recognise faces better than machines.
1:25.5	Hello, today's case is a mystery involving sheep, a car chase and a house that looks like
1:37.1	Hitler. All true. And it starts with this question from the excellently named Karl Vandal,
1:43.0	we wrote to us at Curious Cases at BBC.co.uk to ask our machines better than humans at identifying
1:50.0	faces. This is a classic battle between maths and biology, a face off, if you will. Hold on,
1:56.4	where is the maths in this exactly? It's in the machines, obviously.
2:00.6	Okay, well, now hang on, not so fast, math mode. We humans are simply wired to detect faces
2:06.7	as psychologists Rob Jenkins from the University of York told me. Yeah, we tend to be pretty
2:10.8	good at knowing if there's a face there or not. And that kind of makes sense from an evolutionary
	...

Please login to see the full transcript.

Previous episode | Next episode

Disclaimer: The podcast and artwork embedded on this page are from BBC, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of BBC and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.