Files in this item

FilesDescriptionFormat

application/pdf

application/pdfGULGANJALLINARASIMHAN-THESIS-2019.pdf (13MB)
(no description provided)PDF

Description

Title:Visual question answering using external knowledge
Author(s):Gulganjalli Narasimhan, Medhini
Advisor(s):Schwing, Alexander G.; Lazebnik, Svetlana
Department / Program:Computer Science
Discipline:Computer Science
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:M.S.
Genre:Thesis
Subject(s):Visual question answering, knowledge bases, graph convolution networks
Abstract:Accurately answering a question about a given image requires combining observations with general knowledge. While this is effortless for humans, reasoning with general knowledge remains an algorithmic challenge. To advance research in this direction, a novel `fact-based' visual question answering (FVQA) task has been introduced recently along with a large set of curated facts which link two entities, i.e., two possible answers, via a relation. Given a question-image pair, keyword matching techniques have been employed to successively reduce the large set of facts and were shown to yield compelling results despite being vulnerable to misconceptions due to synonyms and homographs. To overcome these shortcomings, we introduce two new approaches in this work. We develop a learning-based approach which goes straight to the facts via a learned embedding space. We demonstrate state-of-the-art results on the challenging recently introduced factbased visual question answering dataset, outperforming competing methods by more than 5%. Upon further analysis, we observe that a successive process which considers one fact at a time to form a local decision is sub-optimal. To counter this, in our second approach we develop an entity graph and use a graph convolutional network to `reason' about the correct answer by jointly considering all entities. We show on the FVQA dataset that this leads to an improvement in accuracy of around 7% compared to the state-of-the-art.
Issue Date:2019-04-25
Type:Text
URI:http://hdl.handle.net/2142/104918
Rights Information:Copyright 2019 Medhini Gulganjalli Narasimhan
Date Available in IDEALS:2019-08-23
Date Deposited:2019-05


This item appears in the following Collection(s)

Item Statistics