Abstract:
Compact, portable digital cameras have been popular to consumers. This paper presents a robust feature detection method using particle keypoints and its application to vi...Show MoreMetadata
Abstract:
Compact, portable digital cameras have been popular to consumers. This paper presents a robust feature detection method using particle keypoints and its application to video stabilization. The proposed video stabilization algorithm consists of three steps: i) generation of a flat region map based on the Gaussian filter, ii) robust feature point detection using particle keypoints, and iii) camera paths estimation for enhancing a shaky video. As a result, the proposed algorithm can estimate optimal homography by redefining important feature points using particle keypoints in the flat region map. The proposed robust feature detection algorithm is suitable for enhancing the quality of video acquired by consumer handheld imaging devices.
Date of Conference: 07-11 January 2016
Date Added to IEEE Xplore: 14 March 2016
ISBN Information:
Electronic ISSN: 2158-4001