We propose a fast implementation of fractal compression, which uses the intrinsic redundancies within images to decrease file size. Traditional fractal compression is computationally intensive due to the need for extensive comparisons between different parts of the image. Our approach accelerates this process by utilizing Intel's AVX2 intrinsics and optimizing the data locality to minimize cache misses.
Project Collaborators:
- Gianluca Lain
- Stefano Boschetto
- Alessia Paccagnella
- Andrea Ziani