Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/18373
Full metadata record
DC FieldValueLanguage
dc.contributor.authorArif, Arslan-
dc.date.accessioned2021-08-05T19:07:13Z-
dc.date.available2021-08-05T19:07:13Z-
dc.date.issued2019-
dc.identifier.urihttp://hdl.handle.net/123456789/18373-
dc.description.abstractSmart Gallery is an android smartphone photo gallery application that provides visual search option to the user. It is hard for user to find an image from thousands of images in the phone storage. The internal storage of the android smartphones is increasing day by day and one of the main features of latest android phones is mobile photography. People take and share a lot of photos and the smartphone storage is so big that sometimes users cannot find a photo by simple browsing. So the motivation of this project is to provide a visual search option to the users. Users can search images for more than eighty categories of objects so it becomes easy to find a photo. This is all possible due to deep neural networks. The application uses a model trained using deep neural network. The model is imported to android application after being trained on high performance computer. The application then uses the trained model to perform object detection and recognize objects from the images. Thus the user can find any images by just searching an object name like cat, person, car etcen_US
dc.language.isoenen_US
dc.publisherQuaid-i-Azam Universityen_US
dc.subjectComputer Sciencesen_US
dc.titleSmart Galleryen_US
dc.typeThesisen_US
Appears in Collections:BS

Files in This Item:
File Description SizeFormat 
COM 2382.pdfCOM 23821.06 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.