Camera Cognition: Finally, True Facial Recognition

Canon_PowerShot_g1x

Every time I discuss the topics of "facial recognition" and "digital cameras" in the same breath, Jeff Bier validly corrects me. That's because while cameras nowadays commonly do facial detection, i.e. identifying people in a scene and adjusting exposure, focus and other settings to optimize their digital image reproduction, they don't identify specific individuals…

…until now, that is. Thanks to Gizmodo's heads-up, I'm able to point you towards Digital Trends' coverage of this week's Consumer Electronics Show, specifically to a writeup which discusses the capabilities of the DIGIC 5 processor in Canon's new ELPH series and PowerShot G1 X (photo above) cameras. Quoting from Digital Trends' writeup (grammatically edited by yours truly):

One of the most innovative new features is the face ID system. The application is child-weighted, so that younger subjects get preferential treatment when it comes to shooting group shots. There are a host of built-in image recognition tools for taking pictures of babies too, and the cameras will auto-detect and respond accordingly based on how much a baby is smiling and whether or not he’s sleeping (in which case, it won’t use flash).

But these features pale in comparison to its recognition mechanism. Canon told us that using the cameras' DIGIC 5 processor and face recognition technology, you can store up to 12 people. You can add names and birth dates, and take photos at various angles to fill out this profile. And those who you’ve identified will get priority over the nameless.

Check out the video below for more details.

Here you’ll find a wealth of practical technical insights and expert advice to help you bring AI and visual intelligence into your products without flying blind.

Contact

Address

Berkeley Design Technology, Inc.
PO Box #4446
Walnut Creek, CA 94596

Phone
Phone: +1 (925) 954-1411
Scroll to Top