Vision modeling tools for evaluating next-generation displays
Abstract/Contents
- Abstract
- Nearly all metrics for evaluating display image quality operate on the RGB representations used to control conventional flat displays. One computes image quality by comparing an idealized version of the RGB image (reference) and a version rendered by the display under design (test). This framework has been important historically, and it remains suitable for developing metrics that evaluate several aspects of conventional flat display image quality, such as spatial resolution, intensity quantization, and color gamut. There are several limitations of the RGB methods, however. First, standard RGB image quality metrics do not include display characterization, which limits their ability to generalize across ambient viewing conditions, display size, viewing distance, or biological differences between viewers. Second, the RGB representation does not extend to advanced displays where the image is delivered using more complex technology, such as light field, augmented reality, or multi-planar displays. To evaluate the quality of these 3D display technologies, it is necessary to develop a new framework that accommodates the diversity of emerging display technologies. In this dissertation, I adopt a universal representation at the initial stages of the visual pathways by representing display images as retinal photoreceptor (cone) excitations. This is a display-independent representation: all display technologies initiate vision with the retinal images incident on the two eyes. The ISETBio and ISET3d software provides a platform for developing next generation image quality metrics for displays based on cone excitations. The software includes tools for computing not only the retinal image, but also the cone excitations to display images and 3D scenes. In this dissertation, I describe the development of these two toolboxes to model and evaluate next-generation displays that cannot be fully represented by RGB data. First, the existing vision modeling tools were first expanded from 2D image-formation (planar, 2D stimuli) to 3D image-formation (3D virtual models) through the creation and development of ISET3d. Next, the modeled cone excitations from ISETBio/3d were used as input to existing image quality metrics, improving generalization and extending the calculations to novel displays. Lastly, the expanded vision modeling tools were used to model a next-generation, multi-focal display
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2020; ©2020 |
| Publication date | 2020; 2020 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Lian, Trisha Pei-Wei |
|---|---|
| Degree supervisor | Wandell, Brian A |
| Thesis advisor | Wandell, Brian A |
| Thesis advisor | Girod, Bernd |
| Thesis advisor | Wetzstein, Gordon |
| Degree committee member | Girod, Bernd |
| Degree committee member | Wetzstein, Gordon |
| Associated with | Stanford University, Department of Electrical Engineering. |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Trisha Pei-Wei Lian |
|---|---|
| Note | Submitted to the Department of Electrical Engineering |
| Thesis | Thesis Ph.D. Stanford University 2020 |
| Location | electronic resource |
Access conditions
- Copyright
- © 2020 by Trisha Pei-Wei Lian
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...