Object discovery via layer disposal
Name
1076269771-MIT.pdf
Description
Full printable version
Size
11.37 MB
Format
Adobe PDF
Checksum (MD5)
b8cf5d88f8419f3daeccf517e41caae6
Author(s)
Oktay, Deniz, M. Eng. Massachusetts Institute of Technology
Advisor(s)
Antonio Torralba.
Date Issued
2017
Publisher
Massachusetts Institute of Technology
Abstract
A key limitation of semantic image segmentation approaches is that they require large amounts of densely labeled training data. In this thesis, we introduce a method to learn to segment images with unlabeled data. The intuition behind the approach is that removing objects from images will yield natural images, however removing random patches will yield unnatural images. We capitalize on this signal to develop an auto-encoder that decomposes an image into layers, and when all layers are combined, it reconstructs the input image. However, when a layer is removed, the model learns to produce a different image that still looks natural to an adversary, which is possible by removing objects. Experiments and visualizations suggest that this model automatically learns to segment objects in images better than baselines. Some parts of this thesis represent joint work with Dr. Carl Vondrick and Professor Antonio Torralba.
Description
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2017.
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references (pages 51-54).
Subjects
Electrical Engineering and Computer Science.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission.
Persistent DSpace Link
