A magazine where the digital world meets the real world.
On the web
- Home
- Browse by date
- Browse by topic
- Enter the maze
- Follow our blog
- Follow us on Twitter
- Resources for teachers
- Subscribe
In print
What is cs4fn?
- About us
- Contact us
- Partners
- Privacy and cookies
- Copyright and contributions
- Links to other fun sites
- Complete our questionnaire, give us feedback
Search:
Something in the way she moves
How do you tell if the person walking towards you is a man or woman? Getting it wrong can be incredibly embarrassing!
Hairstyle? Men have ponytails. Women have short hair. Body shape? What if they are wearing baggy clothing?
Is there actually any feature that is a reliable way to tell?
Research by Kevin Shan at Queen Mary, University of London has shown that computers can be taught to identify whether a person is male or female, but in a surprising way. When we walk, our limbs act like pendulums, and the length and swing of the pendulum largely determines our individual pattern of movement. This swing is different for boys and girls; it’s due to the difference in our body shapes and skeletons. By showing the computer examples of how lots of people walk the computer can be taught to recognise the difference. To be even better at making the judgment we can combine this with computer facial analysis techniques that are trained to recognise the sorts of shapes of male and female faces. This bringing together of different clues to solve the problem is called 'data fusion'. It turns out it is a key idea in saving the lives of babies too.
See a video of the Queen Mary computer making its identifications...[very big file]
All the clues point to...
Data fusion can go even further combining these and other visual clues such as body temperature, the direction people are looking, or even how long a person has been standing still. From this set of clues a computer can, after it’s been trained and knows the patterns to look for, decide if there is suspicious behaviour going on and alert the appropriate authorities.
So if you have ever been in that embarrassing situation when it dawns on you have been chatting up a woman not the man (or vice versa) you thought you were talking to, at least you know the emotional robots of the future won’t have the same problem.