Accelerating GMM-based patch priors for image restoration: Three ingredients for a 100$\times$ speed-up
This work addresses the practical usability of GMM-based priors for researchers and practitioners in computer vision by making restoration algorithms significantly faster while maintaining competitive performance.
The paper tackled the high runtime complexity of the EPLL algorithm for image restoration by proposing three approximations, achieving a 100x speed-up with less than 0.5 dB drop in quality across tasks like denoising and deblurring.
Image restoration methods aim to recover the underlying clean image from corrupted observations. The Expected Patch Log-likelihood (EPLL) algorithm is a powerful image restoration method that uses a Gaussian mixture model (GMM) prior on the patches of natural images. Although it is very effective for restoring images, its high runtime complexity makes EPLL ill-suited for most practical applications. In this paper, we propose three approximations to the original EPLL algorithm. The resulting algorithm, which we call the fast-EPLL (FEPLL), attains a dramatic speed-up of two orders of magnitude over EPLL while incurring a negligible drop in the restored image quality (less than 0.5 dB). We demonstrate the efficacy and versatility of our algorithm on a number of inverse problems such as denoising, deblurring, super-resolution, inpainting and devignetting. To the best of our knowledge, FEPLL is the first algorithm that can competitively restore a 512x512 pixel image in under 0.5s for all the degradations mentioned above without specialized code optimizations such as CPU parallelization or GPU implementation.