LindaWu - Revision history

imported>Student221: /* Ground truths */

2019-12-15T00:12:29Z

Ground truths

← Older revision		Revision as of 00:12, 15 December 2019
Line 56:		Line 56:

	==Ground truths==		==Ground truths==
	Ground-truth values for image transformations have been used to calculate and demonstrate error in the recovered results with each feature detector and descriptor. For evaluating scale and rotation invariance, ground-truths have been synthetically generated for each image in Dataset-A by resizing and rotating it to known values of scale (50% to 200%) and rotation (1° to 359°). For evaluating the translation invariance, pick the first image in Dataset-B as the ground-truth, and align the ~~reset~~ images to it.		Ground-truth values for image transformations have been used to calculate and demonstrate error in the recovered results with each feature detector and descriptor. For evaluating scale and rotation invariance, ground-truths have been synthetically generated for each image in Dataset-A by resizing and rotating it to known values of scale (50% to 200%) and rotation (1° to 359°). For evaluating the translation invariance, pick the first image in Dataset-B as the ground-truth, and align the rest images to it.

	==Generic image alignment phases==		==Generic image alignment phases==

imported>Student221: /* Experiment of Burst Photography */

2019-12-14T07:07:02Z

Experiment of Burst Photography

← Older revision		Revision as of 07:07, 14 December 2019
Line 267:		Line 267:

	[[File:finalAligned.png\|300px\|link=]]		[[File:finalAligned.png\|300px\|link=]]

	~~The comparison between original image with the enhanced image:~~

	~~[[File:compareAligned.png\|300px\|link=]]~~



Line 277:		Line 272:

	To enhance the original photo, I apply the "averaged" alignment data to the original, which brightens up the regions where were blurred by motion (Step 5).		To enhance the original photo, I apply the "averaged" alignment data to the original, which brightens up the regions where were blurred by motion (Step 5).



Line 293:		Line 289:

	[[File:finalAlginedColor.png\|300px\|link=]]		[[File:finalAlginedColor.png\|300px\|link=]]


			The comparison between original image with the enhanced image:

			[[File:compareAligned.png\|600px\|link=]]

	= Reference =		= Reference =

imported>Student221: /* Experiment of Burst Photography */

2019-12-14T07:04:33Z

Experiment of Burst Photography

← Older revision		Revision as of 07:04, 14 December 2019
Line 267:		Line 267:

	[[File:finalAligned.png\|300px\|link=]]		[[File:finalAligned.png\|300px\|link=]]

			The comparison between original image with the enhanced image:

			[[File:compareAligned.png\|300px\|link=]]

imported>Student221 at 07:03, 14 December 2019

2019-12-14T07:03:15Z

← Older revision		Revision as of 07:03, 14 December 2019
Line 253:		Line 253:
	This computational challenge is connected to the idea of 'burst photography' in which vendors take a series of brief images and then average them to produce a final high-quality image.		This computational challenge is connected to the idea of 'burst photography' in which vendors take a series of brief images and then average them to produce a final high-quality image.

	Experiment 1: Gray-scale Image		'''Experiment 1: Gray-scale Image'''
	*Step 1: Pick any image in the carMovingAway dataset as the original(template) image.		*Step 1: Pick any image in the carMovingAway dataset as the original(template) image.
	*Step 2: Convert it to grayscale using MATLAB function, rgb2gray(rgbOriginal).		*Step 2: Convert it to grayscale using MATLAB function, rgb2gray(rgbOriginal).
Line 274:		Line 274:


	Experiment 2: Color Image		'''Experiment 2: Color Image'''
	*Step 1: Pick any image in the carMovingAway dataset as the original(template) image.		*Step 1: Pick any image in the carMovingAway dataset as the original(template) image.

imported>Student221: /* Experiment of Burst Photography */

2019-12-14T06:55:43Z

Experiment of Burst Photography

@@ Line 253: / Line 253: @@
 This computational challenge is connected to the idea of 'burst photography' in which vendors take a series of brief images and then average them to produce a final high-quality image.
 *Step 1: Pick any image in the carMovingAway dataset as the original(template) image.
 *Step 2: Convert it to grayscale using MATLAB function, rgb2gray(rgbOriginal).
@@ Line 271: / Line 272: @@
 To enhance the original photo, I apply the "averaged" alignment data to the original, which brightens up the regions where were blurred by motion (Step 5).
 = Reference =

imported>Student221 at 06:37, 14 December 2019

2019-12-14T06:37:27Z

← Older revision		Revision as of 06:37, 14 December 2019
Line 258:		Line 258:
	[[File:originalBurst.png\|300px\|link=]]		[[File:originalBurst.png\|300px\|link=]]

	*Step 3: Align the rest of images (the other 7 images) to the template using ORB alignment algorithm. Choosing ORB alignment algorithm ~~becasue~~ it is evaluated as the fastest in this project.		*Step 3: Align the rest of images (the other 7 images) to the template using ORB alignment algorithm. Choosing ORB alignment algorithm because it is evaluated as the fastest in this project.
	*Step 4: Average the aligned images.		*Step 4: Average the aligned images.

imported>Student221: /* Experiment of Burst Photography */

2019-12-13T21:47:02Z

Experiment of Burst Photography

← Older revision		Revision as of 21:47, 13 December 2019
Line 268:		Line 268:


	From the above experiment, Step 4 averaging the alignment data from sequence of burst photos does reduce the noise (misalignment from trees, clouds). To enhance the original photo, I apply the "averaged" alignment data to the original, which brightens up the regions where were blurred by motion (Step 5).		From the above experiment, Step 4 averaging the alignment data from sequence of burst photos does reduce the noise (misalignment from trees, clouds).

			To enhance the original photo, I apply the "averaged" alignment data to the original, which brightens up the regions where were blurred by motion (Step 5).

	= Reference =		= Reference =

imported>Student221: /* Experiment of Burst Photography */

2019-12-13T21:45:51Z

Experiment of Burst Photography

← Older revision		Revision as of 21:45, 13 December 2019
Line 268:		Line 268:


			From the above experiment, Step 4 averaging the alignment data from sequence of burst photos does reduce the noise (misalignment from trees, clouds). To enhance the original photo, I apply the "averaged" alignment data to the original, which brightens up the regions where were blurred by motion (Step 5).

	= Reference =		= Reference =

imported>Student221 at 21:32, 13 December 2019

2019-12-13T21:32:33Z

← Older revision		Revision as of 21:32, 13 December 2019
Line 247:		Line 247:

	The overall accuracy of BRISK and MSER are found to be highest for all types of geometric transformations (as FAST and Harris are not scale invariant), and ORB performs the best with regards to speed versus accuracy.		The overall accuracy of BRISK and MSER are found to be highest for all types of geometric transformations (as FAST and Harris are not scale invariant), and ORB performs the best with regards to speed versus accuracy.

			= Experiment of Burst Photography =
			One of the applications using image alignment algorithm is whether you can align these images and average them!
			The goal is to bring all of the radiance images into register so that the average is less noisy than the originals.
			This computational challenge is connected to the idea of 'burst photography' in which vendors take a series of brief images and then average them to produce a final high-quality image.

			*Step 1: Pick any image in the carMovingAway dataset as the original(template) image.
			*Step 2: Convert it to grayscale using MATLAB function, rgb2gray(rgbOriginal).

			[[File:originalBurst.png\|300px\|link=]]

			*Step 3: Align the rest of images (the other 7 images) to the template using ORB alignment algorithm. Choosing ORB alignment algorithm becasue it is evaluated as the fastest in this project.
			*Step 4: Average the aligned images.

			[[File:averageAlgined.png\|300px\|link=]]

			*Step 5: Add the "averaged aligned data"(Step 4) to the original and produce below.

			[[File:finalAligned.png\|300px\|link=]]



	= Reference =		= Reference =

imported>Student221: /* Dataset */

2019-12-13T20:19:32Z

Dataset

← Older revision		Revision as of 20:19, 13 December 2019
Line 49:		Line 49:

	* '''Dataset-B'''		* '''Dataset-B'''
	The driving scenes generated by ~~iset3D~~ [6] with the camera shifted into multiple positions (see Fig. 2). Currently only translation is involved.		The driving scenes generated by Iset3D [6] with the camera shifted into multiple positions (see Fig. 2). Currently only translation is involved.

	[[Image:carMovingAway.png\|200px\|link=\|caption]] [[Image:drivingScene.png\|200px\|link=]] [[Image:drivingSceneDark.png\|200px\|link=]]		[[Image:carMovingAway.png\|200px\|link=\|caption]] [[Image:drivingScene.png\|200px\|link=]] [[Image:drivingSceneDark.png\|200px\|link=]]