Browsing by Author "Jan, YH"

Now showing 1 - 3 of 3

Edge-and-motion-based semantic video object extraction and application to scene composition
(2004)
Jan, YH
;
Lin, DW
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
We consider automatic segmentation of natural video for content-based video applications. In particular, we present a segmentation algorithm employing both motion and edge information. It employs the edge-linking approach for accurate determination of object boundaries and bidirectional motion estimation for robust tracking of object motion and shape change. The algorithm is designed with computational complexity also in mind. We further consider scene composition using the segmented objects. We, describe and evaluate a method to enlarge or shrink objects for such purpose. The method is amenable to spatial-scalable coding.(1).
Edge-based morphological processing for efficient and accurate video object extraction
(2005-02-01)
Jan, YH
;
Lin, DW
;
電子工程學系及電子研究所
;
電信研究中心
;
Department of Electronics Engineering and Institute of Electronics
We consider the edge-linking approach for accurate locating of moving object boundaries in video segmentation. We review the existing methods and propose a scheme designed for efficiency and better accuracy. The scheme first obtains a very rough outline of an object by a suitable means, e.g., change detection. It then forms a relatively compact image region that properly contains the object, through a procedure termed "mask sketch." Finally, the outermost edges in the region are found and linked via a shortest-path algorithm. Experiments show that the scheme yields good performance.
Image sequence segmentation via heuristic texture analysis and region tracking
(2002)
Jan, YH
;
Lin, DW
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
We develop a method for automatic segmentation of natural video sequences. The method is based on low-level spatial and temporal analyses. It features three designs to help facilitate good region segmentation while keeping the computational complexity at a reasonable level. Firstly, a preliminary seed-area identification and a final re-segmentation process are performed on each video frame to help region tracking. Secondly, a simple way to measure homogeneity of texture in a region is devised and the segmentation tries to locate object boundaries at where the texture shows significant changes. And thirdly, a reduced-complexity motion estimation technique is used, so that dense motion fields can be computed at a reasonable complexity. The overall method is organized into four tasks, namely, seed-area identification (for each frame), initial segmentation (only for the first frame in the sequence), motion-based segmentation (for all later frames), and region tracking and updating (also for all later frames). Some examples are provided to illustrate the performance of this method.