2026-09-19T00:23:24Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1513502023-08-01T03:26:19Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Tenenbaum, Joshua B.
Su, Jocelin
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2023-07-31T19:33:22Z
2023-07-31T19:33:22Z
2023-06
2023-06-06T16:35:03.735Z
https://hdl.handle.net/1721.1/151350
Our visual understanding of the world is factorized and compositional. With just a single observation, we can ascertain both global and local attributes in a scene, such as lighting, weather, and underlying objects. These attributes are highly compositional and can be combined in various ways to create new representations of the world. This paper introduces Decomp Diffusion, an unsupervised method for decomposing images into a set of underlying compositional factors, each represented by a different diffusion model. We demonstrate how each decomposed diffusion model captures a different factor of the scene, ranging from global scene descriptors, (e.g. shadows, foreground, or facial expression) to local scene descriptors (e.g. constituent objects). Furthermore, we show how these inferred factors can be flexibly composed and recombined both within and across different image datasets.
M.Eng.
Massachusetts Institute of Technology
In Copyright - Educational Use Permitted
Copyright retained by author(s)
https://rightsstatements.org/page/InC-EDU/1.0/
Unsupervised Compositional Image Decompositionwith Diffusion Models
Thesis
application/pdf
Master
Master of Engineering in Electrical Engineering and Computer Science
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="af453dbe-c5c9-4d48-a2f6-4e514baa4f1f">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Unsupervised Compositional Image Decompositionwith Diffusion Models</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2023-06</PublicationDate>
<Authors>
<Author>
<DisplayName>Su, Jocelin</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>https://rightsstatements.org/page/InC-EDU/1.0/</License>
<Abstract>Our visual understanding of the world is factorized and compositional. With just a single observation, we can ascertain both global and local attributes in a scene, such as lighting, weather, and underlying objects. These attributes are highly compositional and can be combined in various ways to create new representations of the world. This paper introduces Decomp Diffusion, an unsupervised method for decomposing images into a set of underlying compositional factors, each represented by a different diffusion model. We demonstrate how each decomposed diffusion model captures a different factor of the scene, ranging from global scene descriptors, (e.g. shadows, foreground, or facial expression) to local scene descriptors (e.g. constituent objects). Furthermore, we show how these inferred factors can be flexibly composed and recombined both within and across different image datasets.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>