Semi-automatic video object segmentation
buir.advisor | Onural, Levent | |
dc.contributor.author | Esen, Ersin | |
dc.date.accessioned | 2016-01-08T20:17:10Z | |
dc.date.available | 2016-01-08T20:17:10Z | |
dc.date.issued | 2000 | |
dc.description | Ankara : The Department of Electrical and Electronics Engineering and the Institute of Engineering and Sciences of Bilkent Univ., 2000. | en_US |
dc.description | Thesis (Master's) -- Bilkent University, 2000. | en_US |
dc.description | Includes bibliographical references leaves 70-74 | en_US |
dc.description.abstract | Content-based iunetionalities form the core of the future multimedia applications. The new multimedia standard MPEG-4 provides a new form of interactivity with coded audio-visual data. The emerging standard MPEG-7 specifies a common description of various types of multimedia information to index the data for storage and retrieval. However, none of these standards specifies how to extract the content of the multimedia data. Video object segmentation addresses this task and tries to extract semantic objects from a scene. Two tyj)es of video object segmentation can be identified: unsupervised and supervised. In unsupervised méthods the user is not involved in any step of the process. In supervised methods the user is requested to supply additional information to increase the quality of the segmentation. The proposed weakly supervised still image segmentation asks the user to draw a scribble over what he defines as an object. These scribbles inititate the iterative method. .A.t each iteration the most similar regions are merged until the desired numljer of regions is reached. The proposed .segmentation method is inserted into the unsupervised COST211ter .A-ualysis Model (.A.M) for video object segmentation. The AM is modified to handh' the sujiervision. The new semi-automatic AM requires the user intei actimi for onl>· first frame of the video, then segmentation and object tracking is doin' automatically. The results indicate that the new semi-automatic AM constituK's a good tool for video oliject segmentation. | en_US |
dc.description.provenance | Made available in DSpace on 2016-01-08T20:17:10Z (GMT). No. of bitstreams: 1 1.pdf: 78510 bytes, checksum: d85492f20c2362aa2bcf4aad49380397 (MD5) | en |
dc.description.statementofresponsibility | Esen, Ersin | en_US |
dc.format.extent | xi, 83 leaves, illustrations, tables | en_US |
dc.identifier.uri | http://hdl.handle.net/11693/18199 | |
dc.language.iso | English | en_US |
dc.rights | info:eu-repo/semantics/openAccess | en_US |
dc.subject | Image segmentation | en_US |
dc.subject | Video object segmentation | en_US |
dc.subject | Supervised segmentation | en_US |
dc.subject | Unsupervised segmentation | en_US |
dc.subject | Object tracking | en_US |
dc.subject | MPEG-4 | en_US |
dc.subject | MPEG-7 | en_US |
dc.subject.lcc | TK6680.5 .E84 2000 | en_US |
dc.subject.lcsh | Digitater video. | en_US |
dc.subject.lcsh | Image processing. | en_US |
dc.title | Semi-automatic video object segmentation | en_US |
dc.type | Thesis | en_US |
thesis.degree.discipline | Electrical and Electronic Engineering | |
thesis.degree.grantor | Bilkent University | |
thesis.degree.level | Master's | |
thesis.degree.name | MS (Master of Science) |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- B053306.pdf
- Size:
- 37.11 MB
- Format:
- Adobe Portable Document Format
- Description:
- Full printable version