Fast dejittering for digital video images

Abstract. We propose several very fast algorithms to restore jittered digital video frames (their rows are shifted) in one iteration. The restored row shifts minimize non-smooth and possibly non-convex local criteria applied on the second-order differences between consecutive rows. We introduce specific error measures to assess the quality of dejittering. Our algorithms are designed for gray-value, color and noisy images. Some of them can be considered as parameter-free. They outperform by far the existing algorithms both in quality and in speed. They are a crucial step towards real-time dejittering of digital video.