| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | |
| 7 | 8 | 9 | 10 | 11 | 12 | 13 |
| 14 | 15 | 16 | 17 | 18 | 19 | 20 |
| 21 | 22 | 23 | 24 | 25 | 26 | 27 |
| 28 | 29 | 30 |
« HP's Memory Spot Chip | Main | Researchers develop bistable nanoswitch »
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341ed85853ef00d834bc128f53ef
Listed below are links to weblogs that reference 2D to 3D instant rendering:
This is only a preview. Your comment has not yet been posted.
As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.
Having trouble reading this image? View an alternate.
This is one of those things which appears superficially impressive, but when you dig into the details it turns out to be not quite what it seems.
Reading some of the papers related to this method its revealed that the technique only produces reasonable 3D interpretations about one third of the time (i.e. most images are missinterpreted). The segmentation and vanishing point detection methods used here are really quite brittle, although the basic idea of trying to classify surfaces of different orientations may be useful.
Posted by: Bob Mottram | October 13, 2006 at 07:00 PM