Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Seems to me you could have just had Jetty/OpenCV send back the coords of the faces, then have Javascript draw a box around them. Would have saved quite a bit of bandwidth and removed the need for experimental APIs.


I think you would still want WebRTC and Web Sockets to stream the webcam to Jetty/OpenCV. Yeah, he could have sent just coords back to save on bandwidth though.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: