Show simple item record

dc.contributor.authorWang, Y.
dc.contributor.authorZhu, F.
dc.contributor.authorBoushey, Carol
dc.contributor.authorDelp, E.
dc.date.accessioned2018-05-18T07:58:52Z
dc.date.available2018-05-18T07:58:52Z
dc.date.created2018-05-18T00:23:18Z
dc.date.issued2018
dc.identifier.citationWang, Y. and Zhu, F. and Boushey, C. and Delp, E. 2018. Weakly supervised food image segmentation using class activation maps, pp. 1277-1281.
dc.identifier.urihttp://hdl.handle.net/20.500.11937/67520
dc.identifier.doi10.1109/ICIP.2017.8296487
dc.description.abstract

© 2017 IEEE. Food image segmentation plays a crucial role in image-based dietary assessment and management. Successful methods for object segmentation generally rely on a large amount of labeled data on the pixel level. However, such training data are not yet available for food images and expensive to obtain. In this paper, we describe a weakly supervised convolutional neural network (CNN) which only requires image level annotation. We propose a graph based segmentation method which uses the class activation maps trained on food datasets as a top-down saliency model. We evaluate the proposed method for both classification and segmentation tasks. We achieve competitive classification accuracy compared to the previously reported results.

dc.titleWeakly supervised food image segmentation using class activation maps
dc.typeConference Paper
dcterms.source.volume2017-September
dcterms.source.startPage1277
dcterms.source.endPage1281
dcterms.source.titleProceedings - International Conference on Image Processing, ICIP
dcterms.source.seriesProceedings - International Conference on Image Processing, ICIP
dcterms.source.isbn9781509021758
curtin.departmentSchool of Public Health
curtin.accessStatusFulltext not available


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record