Applying SPIHT on QUINCUNX

[Sumit Kumar (snath@site.uottawa.ca)]

Hi Wael
Unfortunately, SPIHT was devised to be used with a separable wavelet transform. I was also trying to implement a state-of-the-art algorithm with a NS lifting based wavelet transform. Unfortunately, I have not been able to come up with any reasonable solution. The thing that you are looking for, has already be discussed in a paper, oublished in an obscure journal... You can ignore the pathetic lang used in the paper, and instead focus on the essence.
It shows that if you wish to measure image quality with respect to PSNR, the 9-7 separable transform with SPIHT performs much better than its NS counterparts.

www.ktl.elf.stuba.sk/~vargic/papers/radio98/re98_cr.doc

You wont find this journal anywhere . Its supposed to a journal for researchers of the Czech and Slovak republics !