You can convert 2d -> 3d, but its not so easy with just 1 camera.

If you did finish however,  you could get all of you tube and turn it into 3d 
point cloud video.  (so its pretty shmick if you do.)

The only difference between 2 cameras and 1 camera,  is you dont know where the 
next camera position is,   so you can do a brute force search for it.

Do a corner detector,   then find matching corners on the other camera (or next 
frame in the video if its 1 camera) use good old knn for that.

Then you either pre-know the position of the other camera,  or your going to 
run brute force to find it.  (thats the only penalty as i see it for 1 camera, 
or any ordinary video stream.)  

then its just a line-line intersect and u find the z position.

------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T5c987d7a3fa78a4f-Mdf4ef29a016a8d60336e4aa6
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to