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

As you know, to make machine learning work, it needs loads of data.

Most picture and videos are taken from a camera at arm's length, not attached to someone's face.

so if you want to make AR glasses "see" and "understand" the world from the point of view of a human (ie navigation, where is x, etc etc) then you need to make a dataset with that sensor configuration.

https://www.youtube.com/watch?v=6vnZCwf5_QE has a simple example from CMU, rather than facebook



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

Search: