Date of Award

12-31-2018

Document Type

Campus Access Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Lap F. Yu

Second Advisor

Dan Simovici

Third Advisor

wami Iyer

Abstract

Creating 3D interiors is a challenging process. This is especially true with regards to creating a virtual room that matches a certain mood. There is no ensuring that the room will match the mood, since what each person considers a mood could be different. We propose a system for automatically changing the textures for objects within a 3D scene so that the scene matches a certain mood used as input. This is a valuable problem to tackle as it provides users with a tool for matching a mood to an environment with objective criteria. We use datasets of 10,000 images consisting of building/home interiors to train classifiers. This is done for five different moods. These classifiers are then used in an optimization process that scores the mood of a virtual scene. During each step of our optimization process textures of objects are modified until the classifier determines that the room matches the desired input mood. We believe that our approach can provide designers a valuable tool for creating 3D environments and can be used as a tool for studying perceptual design.

Comments

Free and open access to this Campus Access Thesis is made available to the UMass Boston community by ScholarWorks at UMass Boston. Those not on campus and those without a UMass Boston campus username and password may gain access to this thesis through resources like Proquest Dissertations & Theses Global or through Interlibrary Loan. If you have a UMass Boston campus username and password and would like to download this work from off-campus, click on the "Off-Campus UMass Boston Users" link above.

Available for download on Tuesday, December 31, 2019

Share

COinS