Alert icon
We're changing our privacy policy. This stuff matters.  Learn more  Dismiss

Real-time Tracking Based on Object Recognition(II)

Loading...

Sign in or sign up now!
6,608
Loading...
Alert icon
Sign in or sign up now!
Alert icon

Uploaded by on Aug 16, 2007

This video demo illustrates the robustness of the recognition system which is invariant to affine transformation and scales, partial occlusion.

Category:

People & Blogs

Tags:

License:

Standard YouTube License

  • likes, 0 dislikes

Link to this comment:

Share to:

Uploader Comments (ningxu)

  • 中国的?

    能做个交流么?

  • Yes, I am a Chinese student.

  • Have you worked with motion tracking while object consists of high nonlinear transformation and identity? The kernel method will fail if the image is multi-model instead of having global optimal. Do you have other solution?

  • In fact, I've solved multiple object tracking in another way, that is multiple object recognition. You may refer to my other demos in YouTube.

  • this is amazing. i am working on a project to recognize plates. do you know where i could start. amazing job. way the go.

  • I think first you should find a robust object recognition algorithm. I've used Lowe's SIFT(or more accurately, a modification to SIFT) in the demo.

see all

All Comments (6)

Sign In or Sign Up now to post a comment!
  • Looks great! Could you tell me the exact name of the algorithm? Thx!

  • I have seen three of your demostrations. However, it is not exactly the situation I am talking about. If you can take a look at my video samples. I think you may get my point. In my opinion, similarity based tracking algorithm will fail in this case. Maybe we can discuss about this problem sometime?

Loading...
0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more