SIMD parallelization of common wavelet filters
Rade Kutil, Peter J. Eder, M. Watzl · 2005
Much work has been done to optimize wavelet transforms for SIMD extensions of modern CPUs. However, these approaches are always restricted to the vertical part of 2-D transforms with line-wise organized memory layouts because this leads to a rather straight forward SIMDimplementation. This work shows for some common wavelet filters that SIMD operations can also be used on 1-D transforms and produce reasonable speedups. As a result, the performance of algorithms that use wavelet transforms, such as JPEG2000, can be increased significantly. 1