Conference paper
Fast scaled-DCT algorithm
Abstract
The Discrete Cosine Transform (DCT) followed by scaling and quantization is an important operation in image processing. Because of the scaling, the DCT itself need not be computed, but rather a scalar multiple of the DCT might do, with appropriate compensation incorporated into the scaling. We present a fast method for computing such scaled output of the 2-dimensional DCT on 8 × 8 points. We also present a similar algorithm for the inverse scaled DCT.
Related
Conference paper
Unassisted true analog neural network training chip
Conference paper