h a l f b a k e r yThe leaning tower of Piezo
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Look up a table tennis video on youtube.
Despite the best (!?) efforts of the video coding, the small white ball is seen as a detail that doesn't really survive the process of frame coding redundancy. It's dithered or otherwise smoothed out. This is understandable as the ball in many sports is fast
moving and relatively small. It is a prime target, algorithmically, to be removed.
I propose a pre-filtering system that identifies the ball (in any sport) and emphasises it and/or the area around it so that even after the video coding, it remains clear and identifiable.
(Apologies if there are codes out there that do this - I've posted first and asking the question now)
Table Tennis SFX
http://www.youtube....watch?v=-dcmDscwEcI
Some innovative, cutting-edge effects employed here [imaginality, Jul 10 2007]
Joo Se Hyuk vs Werner Schlager
http://www.youtube....watch?v=Oidw8__e6pw
Maybe not so "cutting edge" - but very skilful nonetheless. [Jinbish, Jul 10 2007]
[link]
|
| |
I got told off one time by some Canadians for suggesting that the TV station should highlight the puck in an ice-hockey game, perhaps giving it some tracer type effects, a) to make it look cooler, and b) so I could tell what on earth was going on. |
|
| |
Apparently, some TV company actually did this and it was roundly derided as just not being hockey. |
|
| |
The rumour among some Canadians is that this highlit puck heresy continues to this day in some American coverage of the subsidised violence they call their national game. |
|
| |
I think they do this for hockey broadcasts so that the puck is visible. |
|
| |
They have to do it with hockey. The puck is literally sub pixel in size on an average screen and would simply be eliminated from the picture if not enhanced. |
|
| |
A half pixel black dot in a large white field is going to be treated as an anomaly. |
|
| |
I could see that as more stations appear and more channels update to HD, pressure on bandwidth might push satellite broadcasters into using multiple codecs depending on programming content. Sports programmes could have their own dedicated codec. |
|
| |
Upgrading to mpeg4 would be a start. |
|
| |
Ah - but here is the nub: the encoding is not the problem. It is never the problem. Common digital broadcasting, such as DVB, don't have to use the mpeg2 codecs - they can operate any content information. |
|
| |
It is the hardware that does the decoding that is the problem. More costly and processor intensive receivers would be needed for mpeg4. |
|
| |
That is why I suggest a pre-coding sweep of the pictures. That way the receivers don't have to alter. |
|
| |
A quick and cheap fix. Can it be done? Can you actually predict what an mpeg2 codec is going to do and then reverse engineer the picture to emphasise certain elements? And finally, can you do it? And when? |
|
| |