Natural video synthesis with Generative Adversarial Networks
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Main Author: | |
---|---|
Other Authors: | |
Format: | Thesis |
Language: | eng |
Published: |
Massachusetts Institute of Technology
2019
|
Subjects: | |
Online Access: | https://hdl.handle.net/1721.1/123076 |
_version_ | 1826194025762258944 |
---|---|
author | Egan, Nicholas R.(Nicholas Ryan) |
author2 | Antonio Torralba. |
author_facet | Antonio Torralba. Egan, Nicholas R.(Nicholas Ryan) |
author_sort | Egan, Nicholas R.(Nicholas Ryan) |
collection | MIT |
description | This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. |
first_indexed | 2024-09-23T09:49:21Z |
format | Thesis |
id | mit-1721.1/123076 |
institution | Massachusetts Institute of Technology |
language | eng |
last_indexed | 2024-09-23T09:49:21Z |
publishDate | 2019 |
publisher | Massachusetts Institute of Technology |
record_format | dspace |
spelling | mit-1721.1/1230762020-03-22T04:04:29Z Natural video synthesis with Generative Adversarial Networks Egan, Nicholas R.(Nicholas Ryan) Antonio Torralba. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Electrical Engineering and Computer Science. 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 2019-11-22T00:10:17Z 2019-11-22T00:10:17Z 2019 2019 Thesis https://hdl.handle.net/1721.1/123076 1127639631 eng 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 74 pages application/pdf Massachusetts Institute of Technology |
spellingShingle | Electrical Engineering and Computer Science. Egan, Nicholas R.(Nicholas Ryan) Natural video synthesis with Generative Adversarial Networks |
title | Natural video synthesis with Generative Adversarial Networks |
title_full | Natural video synthesis with Generative Adversarial Networks |
title_fullStr | Natural video synthesis with Generative Adversarial Networks |
title_full_unstemmed | Natural video synthesis with Generative Adversarial Networks |
title_short | Natural video synthesis with Generative Adversarial Networks |
title_sort | natural video synthesis with generative adversarial networks |
topic | Electrical Engineering and Computer Science. |
url | https://hdl.handle.net/1721.1/123076 |
work_keys_str_mv | AT egannicholasrnicholasryan naturalvideosynthesiswithgenerativeadversarialnetworks |