Please use this identifier to cite or link to this item:
https://doi.org/10.1109/ICIP.2007.4380025
DC Field | Value | |
---|---|---|
dc.title | Background cutout with automatic object discovery | |
dc.contributor.author | Liu D. | |
dc.contributor.author | Chen T. | |
dc.date.accessioned | 2018-08-21T05:07:52Z | |
dc.date.available | 2018-08-21T05:07:52Z | |
dc.date.issued | 2007 | |
dc.identifier.citation | Liu D., Chen T. (2007). Background cutout with automatic object discovery. Proceedings - International Conference on Image Processing, ICIP 4 : IV345-IV348. ScholarBank@NUS Repository. https://doi.org/10.1109/ICIP.2007.4380025 | |
dc.identifier.isbn | 1424414377 | |
dc.identifier.isbn | 9781424414376 | |
dc.identifier.issn | 15224880 | |
dc.identifier.uri | http://scholarbank.nus.edu.sg/handle/10635/146279 | |
dc.description.abstract | We present a novel approach to background cutout for image editing. We show how background cutout can be achieved without any user labeling. This is in contrast to current methods, where the user needs to label each image separately. Our method uses automatic object discovery methods to provide location and scale estimates of the object of interest; these estimates then provide seeds for initializing color distributions of a segmentation algorithm. We show that our approach can achieve similar performance as traditional methods that require users to specify for each image a bounding box of the target object. | |
dc.publisher | IEEE Computer Society | |
dc.source | Scopus | |
dc.subject | Image segmentation | |
dc.subject | Unsupervised learning | |
dc.type | Conference Paper | |
dc.contributor.department | OFFICE OF THE PROVOST | |
dc.contributor.department | DEPARTMENT OF COMPUTER SCIENCE | |
dc.description.doi | 10.1109/ICIP.2007.4380025 | |
dc.description.sourcetitle | Proceedings - International Conference on Image Processing, ICIP | |
dc.description.volume | 4 | |
dc.description.page | IV345-IV348 | |
dc.published.state | published | |
Appears in Collections: | Staff Publications |
Show simple item record
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.