Hacker News new | past | comments | ask | show | jobs | submit login

I think what NeilSmithline is trying to say is, the microsoft technology is designed to convert regular 24fps videos to 2.4fps videos (10x time lapse). If you feed it 1 sec timelapse photos (ie, video at 1fps), the output would be 0.1fps.



Yes, it requires the information from all the discarded frames, in order to reconstruct the 3D world for a smooth curve.


I take 1 second time lapse photos and feed them at 60 frames per second for a 60x speed up. I'm not sure if you've made time lapse videos before, but it wouldn't make sense to film something at 1 fps and then play it back at 1 fps.


The frames per second I quote are all recorded frames per second (eg, for every second of real-life time that passes, how many frames in the final video were taken during that second?). Play-back rate is always 24fps (or some other fixed constant).

What we're trying to say is, if you feed your video to the program, you're going to get output that is sped-up 600x compared to real life. That's a ridiculously high speedup.


The difference is that this requires you to record at full speed. With your timelapse you're probably taking a photo every 1 to 5 seconds or so. This system requires you to take 24fps video as your input.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: