Show simple item record

dc.contributor.advisorMostakim, Moin
dc.contributor.authorKamran, Sharif Amit
dc.contributor.authorBin Khaled, Md. Asif
dc.contributor.authorBin Kabir, Sabit
dc.date.accessioned2017-05-09T10:50:39Z
dc.date.available2017-05-09T10:50:39Z
dc.date.copyright2017
dc.date.issued2017-04-19
dc.identifier.otherID 13101176
dc.identifier.otherID 12201105
dc.identifier.otherID 13101194
dc.identifier.urihttp://hdl.handle.net/10361/8112
dc.descriptionThis thesis report is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2017.en_US
dc.descriptionCataloged from PDF version of thesis report.
dc.descriptionIncludes bibliographical references (page 31-33).
dc.description.abstractClassification of images has been a widely regarded challenge for the past decade, but a new type of object recognition problem which deals with pixellevel segmentation is posing a more complex task for both computer vision enthusiasts and researcher alike. The convolutional neural network has become a staple for any recognition task, but a new type of ConvNet which is Fully convolutional in architecture has yielded more fine features and proponents. We propose a neural net where we take VGG19 [20], a well-known classification CNN, make it fully convolutional for extracting deeper features and lastly use skip-architectures[15] for getting finer output. This yields better result than the pre-existing FCN segmentation architecture [15, 25, 6]. Training was done on augmented VOC12 [4] with SBD [6]training data and validation set was used from reduced VOC12 validation dataset. The model scored mIOU of 68.1 percent in PASCAL VOC 2012 Segmentation challenge.en_US
dc.description.statementofresponsibilitySharif Amit Kamran
dc.description.statementofresponsibilityMd. Asif Bin Khaled
dc.description.statementofresponsibilitySabit Bin Kabir
dc.format.extent33 pages
dc.language.isoenen_US
dc.publisherBRAC Universityen_US
dc.rightsBRAC University thesis are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.
dc.subjectNeural networken_US
dc.subjectImage segmentationen_US
dc.titleExploring deep features: deeper fully convolutional neural network for image segmentationen_US
dc.typeThesisen_US
dc.contributor.departmentDepartment of Computer Science and Engineering, BRAC University
dc.description.degreeB. Computer Science and Engineering


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record