Newsmeldung von slashCAM: Hocus Bogus - Super-Resolution of rudi - 11 Aug 2006 13:44:00 > Theoretically it is possible to obtain from standard DV recording noticeably sharper HDV material. The idea behind the "super-resolution" procedures baptized here even sounds quite simple. Just what does all this in practice?
Antwort von Uwe:
In the division is so synonymous of "resizer" of Digital Anarchy:
http://www.digitalanarchy.com/resizer/resizer_main.html
Antwort von Siggis slb:
It is worth while always s.jetzt without (! Film) Tripod too.
HDV of Hand -> Super Resolution -> stabilizer -> Color Correction -> 35mm Faz -> blockbuster -> Villa s.Meer
:)
Antwort von JMitch:
Could it be that the algorithm does not get along with motion blur? Suppose that an object is 2 pixels wide and by the motion blur is 3 pixels wide. Where does the algorithm will know that these are really only 2 pixels? If the motion is constant, then there are always 3 pixels, which he gets to see. The 1st Example, with the license plate looks to me as if the uncertainty is the lack of camera resolution, while it is in your example of motion blur. These synonymous matches, which move in the plant as the plants are also very slow.
An alternative explanation is that the algo does not get along with high speeds. In particular, if the background process of change such as in your car scene.
Markus
Antwort von JMitch:
hmm,
I just found this article and wonder what
in the opposite case, not everything is possible from mediocre to
hdv to sd top recordings recordings, which would have of the
her technical side but be easy!
gruß cj
Antwort von robbie:
Yes, but only if you synonymous TOP - HD - have material. For HD (V) is already so heavily compressed that go much color information lost ...
Schöne Grüße,
Robbie
Antwort von JMitch:
well,
But just as it would have the relationship:
1.55 million pixel hdv (1440x1080) to 0.42 million pixel sd (720x576)
clearly has enough reserves to the missing
information loosely compensate ...
gruß cj
