News for hackers. Stuff that matters.
Re:Yoloface-500k: ultra-light real-time face detection model... by atoav on Wednesday July 08, 2020 @09:40AM (#23769768)
If you are working on technology that detects cancer that is true. If you are working on technology that detects people, this is a harder question.
Re:Yoloface-500k: ultra-light real-time face detection model... by 0-_-0 on Wednesday July 08, 2020 @09:59AM (#23769908)
The same neural network can be trained for both problems. Advances in medical image neural networks translate to advances in people detecting neural networks.
Re:Yoloface-500k: ultra-light real-time face detection model... by joshuamorton on Wednesday July 08, 2020 @11:57AM (#23771135)
Generally speaking this isn't super true, the objects are different enough that the same networks aren't best at both.
Re:Yoloface-500k: ultra-light real-time face detection model... by woko on Thursday July 09, 2020 @09:29AM (#23779932)
Of course, they are not directly used without changing the training data, etc. However, if we look at the big picture to make a general statement, then the advances brought by Computer Vision on natural images often help push research on medical images forward, albeit with a year (or two) of delay. One such example is U-Net (https://en.wikipedia.org/wiki/U-Net).