SoC2007 project Anti Ghosting - Revision history

Flixh: Undo revision 12352 by Marie101, revert link spamming

2010-05-04T20:52:18Z

Undo revision 12352 by Marie101, revert link spamming

Marie101 at 11:03, 4 May 2010

2010-05-04T11:03:39Z

Jingidy: /* Deliverables and Details */

2007-07-13T05:50:49Z

‎Deliverables and Details

Jingidy at 06:03, 14 April 2007

2007-04-14T06:03:48Z

Erik Krause: categorized

2007-03-27T19:54:24Z

categorized

Jingidy at 15:56, 27 March 2007

2007-03-27T15:56:28Z

Jingidy at 14:15, 27 March 2007

2007-03-27T14:15:37Z

New page

==Goal==

Most HDR creation algorithms are designed to work only with very small variations in camera viewing direction. Assume that registration and response curve estimation has already happened. An improved blending method for HDR images together that have not been shot using the traditional exposure stack method. It should avoid ghosting and be insensitive to small misregistrations.

==Required knowledge or interest==

* Strong background signal/image processing
* Creative mind with ideas beyond the state of the art in computer vision/graphics research.

==Deliverables and Details==

==Schedule==

==Mentor==

==Students planning to apply==

[[User:jingidy | Jing Jin]]

==Resources==

← Older revision		Revision as of 20:52, 4 May 2010
Line 54:		Line 54:

	==Resources==		==Resources==
−	~~want to spend a vacation with your family? treat them like you've never done before, make then experience what is a real vacation place, then you might want to try this site.~~
−
−	~~Related Link:~~
−
−	~~[http://www.fraserislandtours.com.au/sites/default.asp '''fraser island tours from hervey bay''']~~

	* [http://www.cs.berkeley.edu/~eden/737_eden_a.pdf Seamless Image Stitching of Scenes with Large Motions and Exposure Differences]		* [http://www.cs.berkeley.edu/~eden/737_eden_a.pdf Seamless Image Stitching of Scenes with Large Motions and Exposure Differences]

← Older revision		Revision as of 11:03, 4 May 2010
Line 54:		Line 54:

	==Resources==		==Resources==
		+	want to spend a vacation with your family? treat them like you've never done before, make then experience what is a real vacation place, then you might want to try this site.
		+
		+	Related Link:
		+
		+	[http://www.fraserislandtours.com.au/sites/default.asp '''fraser island tours from hervey bay''']

	* [http://www.cs.berkeley.edu/~eden/737_eden_a.pdf Seamless Image Stitching of Scenes with Large Motions and Exposure Differences]		* [http://www.cs.berkeley.edu/~eden/737_eden_a.pdf Seamless Image Stitching of Scenes with Large Motions and Exposure Differences]

← Older revision		Revision as of 05:50, 13 July 2007
Line 6:		Line 6:

	==Deliverables and Details==		==Deliverables and Details==
		+	Go to [http://hugindeghosting.blogspot.com/ the blog] for project progress, results, and more!

	==Schedule==		==Schedule==

@@ Line 23: / Line 23: @@
 Phase 1 - Research (2 weeks)
-* Week 1: research different algorithms used in current softwares and proposed in research papers. Understand  their pros and cons.
+* Weeks 1-2:
-* Week2: design an efficient and robust algorithm and plan out the detailed schedule of the implementation and integration phases. Also design the main classes/functions that are involved in the algorithm and those that are needed to integrate.
+** obtain several sets of sample images to use for testing later
 Phase 2 - Implementation (7 weeks)
-* Weeks 3-6 implement the merging algorithm and do any unit and incremental tests involved
+* Week 3
 * Weeks 7-9 time for general testing and debugging
 Phase 3 - Integration (5 weeks)
-* Weeks 10-11 integrate the algorithm with the rest of the panorama stitching process, testing as needed/possible
+* Weeks 10 integration into hugin
-* Weeks 12-13 robust testing debugging of the whole process
+* Weeks 11-13 testing/debugging and optimization
-* Week 14 optimization, documentation, etc.
+* Week 14 documentation, etc.
 ==Mentors==
-Pablo d'Angelo, Yuval Levy
+Pablo d'Angelo
 ==Students planning to apply==

@@ Line 1: / Line 1: @@
 ==Goal==
@@ Line 46: / Line 48: @@
 * [http://graphics.cs.ucf.edu/ekhan/project_ghost.htm Ghost Removal in High Dynamic Range Images]
 * [http://research.microsoft.com/IVM/HDR/hdr_sg2003.pdf High Dynamic Range Video]

@@ Line 1: / Line 1: @@
 ==Goal==
-Most HDR creation algorithms are designed to work only with very small variations in camera viewing direction. Assume that registration and response curve estimation has already happened. An improved blending method for HDR images together that have not been shot using the traditional exposure stack method. It should avoid ghosting and be insensitive to small misregistrations.
+High Dynamic Range (HDR) Panoramas are formed by taking numerous pictures of different exposures, tone mapping them to an exposure that provides the most detail for the human eye, and stitching them into a single image. Since there are multiple base images per area in HDR imags, moving objects, such as people, result in semi-transparency, blurring, or incomplete objects. This phenomenon is called ghosting. Some commercial products support ghosting elimination, but most of them only work with small variations in the camera. The goal of this project is to devise a robust blending algorithm to eliminate ghosting in an HDR panorama.
-−
-==Required knowledge or interest==
-−
-*  Strong background signal/image processing
-* Creative mind with ideas beyond the state of the art in computer vision/graphics research.
 ==Deliverables and Details==
@@ Line 12: / Line 7: @@
 ==Schedule==
-==Mentor==
+The execution of the project involves 3 phases: research, implementation, and integration.
 ==Students planning to apply==
@@ Line 19: / Line 42: @@
 ==Resources==