Digital image stabilization by adaptive block motion vectors filtering

Abstract

This paper presents a robust algorithm for video sequences stabilization. Motion estimation is achieved using block motion vectors. In this way the same motion estimator of mpeg encoder can be used. The simple use of block motion vectors can give unreliable global motion vectors and so elaborations are done to make the algorithm robust.

Fast Image Stitching For Video Stabilization Using Sift Feature Points

“…To do so, at first minimum displacement according to x-axis and y-axis (x_min_dis, y_min_dist) from current frame to next frame is calculated at every iteration. Now for every initial matched keypoints the displacement in x axis (|x1-x2|) and y axis (|y1-y2|) are calculated and checked with the equation (4). Assuming all the keypoints in initial match have minimum displacement then total average minimum displacement for all initial match would be (initial_match_size* (min_dist_x+min_dist_y)).…”

“…In early days, 'block matching' was used to estimate local motion and remove unwanted motions [1][2][3][4] .…”

Comparison of Motion Smoothing Strategies for Video Stabilization using Parametric Models

“…Motion estimation: Classic methods [20,45,46,60] either use correlation techniques or track salient points such as Harris corners [21] using, e.g., a KLT tracker [42,51]. Pyramidal structures are typically used for estimating large displacements [43,18,48].…”

Automatic Feature-Based Stabilization of Video with Intentional Motion through a Particle Filter

“…Different techniques have been proposed for digital video stabilization, which differ in the method they use to compute the image motion. Block-matching [1] [2] methods divide a frame into blocks, and compute a motion vector for each one through the searching of the more similar block in the next frame. However, the motion estimation could be biased in low-textured image regions due to the aperture problem [3].…”