Sign up Calendar Latest Topics
 
 
 


Reply
  Author   Comment  
Jaewon

Administrator
Registered:
Posts: 716
Reply with quote  #1 
Hi all,

I have a poster presentation on Monday morning (325.01, DDD14). My poster is not about ML this year, but you can come by if you have ML issues to discuss.

I am currently working on video capturing in ML. It is almost done, but I didn't figure out yet the best way to handle huge video data. Please let me know if you are interested in video capture or have any suggestion.
0
Edward

Administrator
Registered:
Posts: 245
Reply with quote  #2 
Hey Jaewon,

I have done a few projects in matlab before. I also have some experience with mex files. I'd be interested to learn more about your upcoming video capturing feature in more technical details. 
0
Jaewon

Administrator
Registered:
Posts: 716
Reply with quote  #3 
Video capturing itself is not a problem and is finished already. The problem is that, with video capturing, the size of data produced per hour is tens of gigabytes, instead of tens of megabytes. I am trying to reduce it to the level that we can practically handle.
0
Edward

Administrator
Registered:
Posts: 245
Reply with quote  #4 
Hmm, if you need to compress the data from tens of gigabytes down to tens of megabytes (0.1% compression), then in addition to saving the video in lower quality compressed format, it will be necessary to skip quite a few non important frames. Does the captured video have any characteristics that can be used to throw away, say 90% of the captured frames?
0
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.