UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Assistive visual content creation tools via multimodal correlation analysis

Hennessey, James W.; (2018) Assistive visual content creation tools via multimodal correlation analysis. Doctoral thesis (Eng.D), UCL (University College London). Green open access

[thumbnail of james_hennessey_engd_thesis_compressed.compressed.pdf]
Preview
Text
james_hennessey_engd_thesis_compressed.compressed.pdf - Accepted Version

Download (9MB) | Preview

Abstract

Visual imagery is ubiquitous in society and can take various formats: from 2D sketches and photographs to photorealistic 3D renderings and animations. The creation processes for each of these mediums have their own unique challenges and methodologies that artists need to overcome and master. For example, for an artist to depict a 3D scene in a 2D drawing they need to understand foreshortening effects to position and scale objects accurately on the page; or, when modeling 3D scenes, artists need to understand how light interacts with objects and materials, to achieve a desired appearance. Many of these tasks can be complex, time-consuming, and repetitive for content creators. The goal of this thesis is to develop tools to alleviate artists from some of these issues and to assist them in the creation process. The key hypothesis is that understanding the relationships between multiple signals present in the scene being created enables such assistive tools. This thesis proposes three assistive tools. First, we present an image degradation model for depth-augmented image editing to help evaluate the quality of the image manipulation. Second, we address the problem of teaching novices to draw objects accurately by automatically generating easy-to-follow sketching tutorials for arbitrary 3D objects. Finally, we propose a method to automatically transfer 2D parametric user edits made to rendered 3D scenes to global variations of the original scene.

Type: Thesis (Doctoral)
Qualification: Eng.D
Title: Assistive visual content creation tools via multimodal correlation analysis
Event: University College London
Open access status: An open access version is available from UCL Discovery
Language: English
UCL classification: UCL
UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
URI: https://discovery.ucl.ac.uk/id/eprint/10046053
Downloads since deposit
243Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item