Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>This is a difficult problem and any solution will not be perfect. Computer vision is jokingly known as an "AI-complete" discipline: if you solve computer vision and you have solved all of artificial intelligence.</p> <p>Background subtraction can be a good way of detecting objects. If you need to improve the background subtraction results, you might consider using an <a href="https://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=4129585" rel="nofollow noreferrer">MRF</a>. Presumably, you can tell when there is a single object and when the two blobs have merged, based on the size of the blob. If the trajectories don't change quickly during the times the blobs are merged, you can do <a href="http://www.cs.unc.edu/~tracker/media/pdf/SIGGRAPH2001_CoursePack_08.pdf" rel="nofollow noreferrer">Kalman tracking</a> and use some <a href="http://people.csail.mit.edu/cielbleu/pubs/BoseEtalCVPR07Multiclass.pdf" rel="nofollow noreferrer">heuristics</a> to disambiguate the blobs afterwards.</p> <p>Even though the colors are similar between the two objects, you might consider trying to use a <a href="http://ieeexplore.ieee.org/search/wrapper.jsp?arnumber=1000236" rel="nofollow noreferrer">mean shift tracker</a>. It's possible that you may need to do some <a href="https://en.wikipedia.org/wiki/Particle_filter" rel="nofollow noreferrer">particle filtering</a> to keep track of multiple hypotheses about who is who.</p> <p>There are also some even more complicated techniques called <a href="http://persci.mit.edu/demos/jwang/garden-layer/layer-demo.html" rel="nofollow noreferrer">layered tracking</a>. There is some more recent work by <a href="http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.11.7433" rel="nofollow noreferrer">Jojic and Frey</a>, by <a href="http://johnwinn.org/Research/Affine_tracking.html" rel="nofollow noreferrer">Winn</a>, by <a href="http://lear.inrialpes.fr/people/triggs/events/iccv03/cdrom/iccv03/1079_zhou.pdf" rel="nofollow noreferrer">Zhou and Tao</a>, and by others. Most of these techniques come with very strong assumptions and/or take a lot of work to implement correctly.</p> <p>If you're interested in this topic in general, I highly recommend taking a computer vision course and/or reading a textbook such as <a href="http://rads.stackoverflow.com/amzn/click/0130851981" rel="nofollow noreferrer">Ponce and Forsyth's</a>.</p>
    singulars
    1. This table or related slice is empty.
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. This table or related slice is empty.
    2. VO
      singulars
      1. This table or related slice is empty.
    3. VO
      singulars
      1. This table or related slice is empty.
 

Querying!

 
Guidance

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload