- OpenBU
- Theses & Dissertations
- Boston University Theses & Dissertations
Manipulating natural images by learning relationships between visual domains
Date Issued
Export Citation
Permanent link, collections.
- Boston University Theses & Dissertations [9206]
Deposit Materials
University of Illinois Urbana-Champaign
Making image generation and manipulation simple and effective
Chong, min jin.
https://hdl.handle.net/2142/117765 Copy
Description
- Schwing, Alexander
- Hoiem, Derek
- Wang, Yuxiong
- Fidler, Sanja
- Generative Models
- Generative Adversarial Networks
- Stylization
- Image Manipulation
- Image Generation
- Accelerating GANs
Owning Collections
Graduate dissertations and theses at illinois primary, manage files, edit collection membership, edit metadata, edit properties.
I don't have an Illinois NetID
COMMENTS
In this thesis, we investigate whether flexible attribute manipulation models can be trained without massive labeled datasets of real images by transferring knowledge about the desired manipulation across different image datasets (domains) that share the underlying structure.
In this thesis, we explore techniques that allow casual users to perform complex semantic image manipulation with ease in an intuitive way. Enabling such interactions requires a deep knowledge of images and their underlying structures.
I describe the various types of image manipulation, giving examples, and then canvas the literature to describe the landscape of image manipulation problems and extant solutions, namely: the nature of image manipulation, investigations of human perceptions of image manipulation, eye gaze tracking and manipulated images,