Introduction - If you have any usage issues, please Google them yourself
HSF processes the distorted image ˜ x by M filters chosen to improve the image quality for regions with different characteristics. At each pixel,
a feature vector yn is computed using local image data. The feature vector is then used to determine the optimal filter weights to form the final output.
The mapping between the feature vector and the optimal filter weights is controlled by a number of model parameters which are determined through
training.