UIUC CS 498 - Computational Photography - Final project presentation
Download as PPT, PDF1 like962 views
This document outlines a process for simulating miniature photography effects through computational techniques. It discusses using depth of field specification and blurring to create a miniature effect from a standard photo. The author presents several methods for specifying depth of field, including using horizontal focus lines, object masks, and salient region detection. Open issues discussed are depth estimation from a single image and improving salient region selection to better automate the miniature photography simulation process.
![Computational Photography -Final Project- Miniature Photography Jia-Bin Huang [email_address]](https://image.slidesharecdn.com/finalprojectpresentation-101220171659-phpapp01/85/UIUC-CS-498-Computational-Photography-Final-project-presentation-1-320.jpg)
![References [1] Held et al. " Using Blur to Affect Perceived Distance and Size ". ACM Transactions on Graphics 2010. [2] Levin et al. " Image and Depth from a Conventional Camera with a Coded Aperture ". SIGGRAPH 2007. [3] Felzenszwalb and Huttenlocher. " Efficient Graph-Based Image Segmentation ". IJCV 2004. [4] Achant et al. " Frequency-tuned Salient Region Detection ". CVPR 2009.](https://image.slidesharecdn.com/finalprojectpresentation-101220171659-phpapp01/85/UIUC-CS-498-Computational-Photography-Final-project-presentation-24-320.jpg)
UIUC CS 498 - Computational Photography - Final project presentation
- 1. Computational Photography -Final Project- Miniature Photography Jia-Bin Huang [email_address]
- 4. Outline Introduction of fake blur Process Overview Depth-of-Field (DOF) Specification Results Discussion
- 5. Introduction of Fake Miniature Optical blur Slide credit: Rebert Held
- 6. Normal Lens Slide credit: Rebert Held
- 7. Tilt-Shift Lens Slide credit: Rebert Held
- 8. Process Overview Goal: Simulate the Tilt-Shift Effect (i.e., miniature photography) Process Focus line/region/target selection DOF specification Blur image Color Enhancement (optional)
- 9. Depth-of-Field (DOF) Specification We lost depth information from the image. Assumption: objects on the same straight line are at the same depth. Methods to specify DOF Horizontal Focus line Object mask Salient region detection
- 10. DOF - Horizontal
- 11. DOF - Focus Line
- 12. DOF – Object Mask
- 13. DOF – Salient Region Detection
- 14. More Results
- 23. Discussion Two main open issues Depth estimation from single image Salient region selection Accurate depth information can help create realistic miniature effect Salient region selection can help automate the miniature photography process
- 24. References [1] Held et al. " Using Blur to Affect Perceived Distance and Size ". ACM Transactions on Graphics 2010. [2] Levin et al. " Image and Depth from a Conventional Camera with a Coded Aperture ". SIGGRAPH 2007. [3] Felzenszwalb and Huttenlocher. " Efficient Graph-Based Image Segmentation ". IJCV 2004. [4] Achant et al. " Frequency-tuned Salient Region Detection ". CVPR 2009.