2026-09-18T21:57:17Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1230762026-06-06T00:49:11Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Antonio Torralba.
Egan, Nicholas R.(Nicholas Ryan)
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2019-11-22T00:10:17Z
2019-11-22T00:10:17Z
2019
2019
https://hdl.handle.net/1721.1/123076
1127639631
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2019
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 71-74).
Generative Adversarial Networks (GANs) are the state of the art neural network models for image generation, but the use of GANs for video generation is still largely unexplored. This thesis introduces new GAN based video generation methods by proposing the technique of model inflation and the segmentation-to-video task. The model inflation technique converts image generative models into video generative models, and experiments show that model inflation improves training speed, training stability, and output video quality. The segmentation-to-video task is that of turning an input image segmentation mask into an output video matching that segmentation. A GAN model was created to perform this task, and its usefulness as a creative tool was demonstrated.
by Nicholas R. Egan.
M.Eng.
M.Eng. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science
74 pages
eng
Massachusetts Institute of Technology
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.
http://dspace.mit.edu/handle/1721.1/7582
Electrical Engineering and Computer Science.
Natural video synthesis with Generative Adversarial Networks
Thesis
application/pdf
2020-03-09T19:58:08Z
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="32e817c3-aff8-4f47-a7fa-e6927282d191">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>eng</Language>
<Title>Natural video synthesis with Generative Adversarial Networks</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2019</PublicationDate>
<Authors>
<Author>
<DisplayName>Egan, Nicholas R.(Nicholas Ryan)</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://dspace.mit.edu/handle/1721.1/7582</License>
<Keyword>Electrical Engineering and Computer Science.</Keyword>
<Abstract>Generative Adversarial Networks (GANs) are the state of the art neural network models for image generation, but the use of GANs for video generation is still largely unexplored. This thesis introduces new GAN based video generation methods by proposing the technique of model inflation and the segmentation-to-video task. The model inflation technique converts image generative models into video generative models, and experiments show that model inflation improves training speed, training stability, and output video quality. The segmentation-to-video task is that of turning an input image segmentation mask into an output video matching that segmentation. A GAN model was created to perform this task, and its usefulness as a creative tool was demonstrated.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>