Gyro-based multi-image deconvolution for removing handshake blur
Abstract/Contents
- Abstract
- In low-light situations, long exposures are inevitable to improve the image quality, but, in turn, handshake blur becomes significant. Although image deblurring techniques have been intensively studied, most methods are brittle and computationally expensive, making them hardly usable in practice. This dissertation presents a new gyro-based multi-image deblurring system that handles various challenges in real-world situations. Instead of capturing a single long exposure image, a burst of shorter exposure images is captured along with gyroscope data being recorded at the same time. Then, any camera shake that occurred during the capture is removed to obtain a sharp and noise-free image of the scene. Because both camera motion and the scene are accurately measured with a better capture method, gyro-based multi-image deblurring is more robust than previous approaches while its computational burden is significantly reduced. In addition to the implementation, an important property of multi-image deblurring is presented: Multi-image deconvolution does not benefit from increasing exposure time beyond a certain threshold. The noise reduction granted by increasing exposure time is mostly offset by the noise amplification incurred by the more extreme deblurring. Based on this analysis, different capture strategies are compared in various photographic situations, demonstrating that multi-image deblurring is a robust approach that works in low-light situations when other methods fail.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2014 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Park, Sung-Hee |
|---|---|
| Associated with | Stanford University, Department of Electrical Engineering. |
| Primary advisor | Levoy, Marc |
| Thesis advisor | Levoy, Marc |
| Thesis advisor | Horowitz, Mark (Mark Alan) |
| Thesis advisor | Rock, Stephen M |
| Advisor | Horowitz, Mark (Mark Alan) |
| Advisor | Rock, Stephen M |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Sung Hee Park. |
|---|---|
| Note | Submitted to the Department of Electrical Engineering. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2014. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2014 by Sung Hee Park
- 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...