Please use this identifier to cite or link to this item: http://repository.iiitd.edu.in/xmlui/handle/123456789/894
Title: Image repurposing detection
Authors: Kumar, Ashwat
Shah, Rajiv Ratn (Advisor)
Keywords: Multimedia Verification, Information Retrieval, Multimodal Representation Learning, Outlier Detection, Natural Language Processing
Issue Date: 1-Jun-2020
Publisher: IIIT-Delhi
Abstract: Image Repurposing is a type multimedia manipulation in which instead of manipulating image and it’s content, manipulation is done on it’s associated modality. For example - image’s caption, location metadata, author metadata etc. You can change the purpose/motive of original content, by publishing incorrect/false data with unmanipulated image. I aim to build a framework on which you can query and get result whether query has been repurposed or is in original form. I have modelled this problem as verification problem. We have a reference database to verify upon. The overall pipeline consists of first retrieving relevant results, checking relevance/relationship with query and finally checking inlierness of query with the retrieved results or whether query is outlier or not
URI: http://repository.iiitd.edu.in/xmlui/handle/123456789/894
Appears in Collections:Year-2020

Files in This Item:
File Description SizeFormat 
Ashwat Kumar-2016023.pdf
  Restricted Access
2.04 MBAdobe PDFView/Open Request a copy


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