I’m stuck ! I’ve been wanting to add color to the video-chat test app, but I can’t seem to focus and decide how to do it.
I want to do something clever. I’m already compressing the luminance (gray level image) at relatively high quality. To compress the remaining two color components without utilizing the similarities with the luminance it would be a crime. At the same time, it’s not easy to just write code for an algorithm without even a source of inspiration.
If I don’t come up with something clever, I’m bound to force the data size to sensibly grow, or have poor quality. This doesn’t particularly make me happy. On the other hand, black and white images really suck 8)
The truth is that I should sit down and start making experiments. Compression is all about statistics and statistics is all about inspecting data. In all honesty, I’m not sure how chrominance components are distribuited, therefore I can’t really come up with something good.
Doubt is probably the number one killer for productivity !
Posted by Davide Pasca in Uncategorized
