<p>Of course it would. :)</p>
<p>Computationally you could do all in GPU and track everything fast as hell . . . Or parallelize chunks of code into a multicore, and so forth. </p>
<p>But I don't really care. Tracking blobs with OpenCV takes me about 8ms (VGA size camera) in a normal computer, and without a threaded approach, so perfect, suits me without worries, i can have just stich two cams and it gives about 14ms, which is more than great for me. </p>
<p>On May 24, 2011 7:34 AM, "IOhannes m zmoelnig" <<a href="mailto:zmoelnig@iem.at">zmoelnig@iem.at</a>> wrote:<br type="attribution">> -----BEGIN PGP SIGNED MESSAGE-----<br>> Hash: SHA1<br>> <br>
> On 2011-05-23 19:55, Pedro Lopes wrote:<br>>> It seems just: image blending. Each camera sees a portion (quarter) of the<br>>> floor, and you blend (or stitch) the images, then supply them to the<br>>> tracking algorithm as single image.<br>
>> <br>> <br>> computationally more efficient could be to do tracking on the 4 separate<br>> images and then "stitch" together the resulting data.<br>> <br>> fgmasdr<br>> IOhannes<br>> -----BEGIN PGP SIGNATURE-----<br>
> Version: GnuPG v1.4.11 (GNU/Linux)<br>> Comment: Using GnuPG with Mozilla - <a href="http://enigmail.mozdev.org/">http://enigmail.mozdev.org/</a><br>> <br>> iEYEARECAAYFAk3bXlYACgkQkX2Xpv6ydvRmogCfWz4eOymxzS5t3MxgJKuk6IyC<br>
> QWQAnjR3g94Wv7swKzjUZv6ANDx6XuQT<br>> =SPVp<br>> -----END PGP SIGNATURE-----<br>> <br></p>