CollinsZvinakisDanowitz - Revision history

imported>Psych2012: /* Amazon Mechanical Turk observer ratings on retouching severity */

2012-03-22T07:35:57Z

Amazon Mechanical Turk observer ratings on retouching severity

← Older revision		Revision as of 07:35, 22 March 2012
Line 210:		Line 210:

	1. Each participant was initially shown a representative set of 20 before/after images to help them gauge the range of distortions that they could expect to see		1. Each participant was initially shown a representative set of 20 before/after images to help them gauge the range of distortions that they could expect to see

	2. Each participant was then asked to rate 70 pairs of before/after images on a scale of 1 (“very similar”) to 5 (“very different”). The presentation of images was self-timed; participants could manually toggle between before and after images as many times as they wished. Each observer rated a random set of 5 images 3 times each to measure the consistency of their responses.		2. Each participant was then asked to rate 70 pairs of before/after images on a scale of 1 (“very similar”) to 5 (“very different”). The presentation of images was self-timed; participants could manually toggle between before and after images as many times as they wished. Each observer rated a random set of 5 images 3 times each to measure the consistency of their responses.

imported>Psych2012: /* Photometric model */

2012-03-22T07:32:04Z

Photometric model

← Older revision		Revision as of 07:32, 22 March 2012
Line 52:		Line 52:
	The local linear filter is used to determine the extent of the sharpening or blurring that occurs over a small region of the image. This computation is done by estimating a matrix “H” that transforms a portion of the registered, unaltered image into the corresponding part of the altered image, as shown in equation 2. Since different parts of the image may be sharpened and blurred independently, H must be computed on local regions only to determine the extent of photometric image alterations. For our reimplementation of the algorithm, we analyzed 4x4 pixel blocks of the images.		The local linear filter is used to determine the extent of the sharpening or blurring that occurs over a small region of the image. This computation is done by estimating a matrix “H” that transforms a portion of the registered, unaltered image into the corresponding part of the altered image, as shown in equation 2. Since different parts of the image may be sharpened and blurred independently, H must be computed on local regions only to determine the extent of photometric image alterations. For our reimplementation of the algorithm, we analyzed 4x4 pixel blocks of the images.

	<math>F_{after}=HF_{before}</math> [2]		<math>F_{after}=HF_{before}</math> (2)

	We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.		We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.

	<math>D=\sum\|F_{before}(\omega)\|\omega-\sum\|HF_{before}(\omega)\|\omega</math> [3]		<math>D=\sum\|F_{before}(\omega)\|\omega-\sum\|HF_{before}(\omega)\|\omega</math> (3)

	We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.		We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.

imported>Psych2012 at 07:31, 22 March 2012

2012-03-22T07:31:04Z

← Older revision		Revision as of 07:31, 22 March 2012
Line 180:		Line 180:

	= References - Resources and related work =		= References - Resources and related work =
	Farid’s image transformation MATLAB code:		1. Farid’s image transformation MATLAB code:
	http://www.cs.dartmouth.edu/farid/Hany_Farid/Research/Entries/2011/5/17_Image_Registration.html		http://www.cs.dartmouth.edu/farid/Hany_Farid/Research/Entries/2011/5/17_Image_Registration.html
	CVX: http://cvxr.com/cvx/		2. CVX: http://cvxr.com/cvx/
	Zhou Wang’s SSIM code: https://ece.uwaterloo.ca/~z70wang/research/ssim/		3. Zhou Wang’s SSIM code: https://ece.uwaterloo.ca/~z70wang/research/ssim/
	Implementation (or SSIM in general) described in this paper:		4. Implementation (or SSIM in general) described in this paper:
	http://www.cns.nyu.edu/pub/eero/wang03-reprint.pdf		http://www.cns.nyu.edu/pub/eero/wang03-reprint.pdf
	libSVM: http://www.csie.ntu.edu.tw/~cjlin/libsvm/		5. libSVM: http://www.csie.ntu.edu.tw/~cjlin/libsvm/
	Parameter selection in MATLAB: http://www.csie.ntu.edu.tw/~cjlin/libsvm/faq.html#f803		6. Parameter selection in MATLAB: http://www.csie.ntu.edu.tw/~cjlin/libsvm/faq.html#f803
	John Brunhaver wrote the two main functions in the photo_batch		7. John Brunhaver wrote the two main functions in the photo_batch

	= Appendix I - Code and Data =		= Appendix I - Code and Data =

imported>Psych2012: /* Photometric model */

2012-03-22T07:30:07Z

Photometric model

← Older revision		Revision as of 07:30, 22 March 2012
Line 52:		Line 52:
	The local linear filter is used to determine the extent of the sharpening or blurring that occurs over a small region of the image. This computation is done by estimating a matrix “H” that transforms a portion of the registered, unaltered image into the corresponding part of the altered image, as shown in equation 2. Since different parts of the image may be sharpened and blurred independently, H must be computed on local regions only to determine the extent of photometric image alterations. For our reimplementation of the algorithm, we analyzed 4x4 pixel blocks of the images.		The local linear filter is used to determine the extent of the sharpening or blurring that occurs over a small region of the image. This computation is done by estimating a matrix “H” that transforms a portion of the registered, unaltered image into the corresponding part of the altered image, as shown in equation 2. Since different parts of the image may be sharpened and blurred independently, H must be computed on local regions only to determine the extent of photometric image alterations. For our reimplementation of the algorithm, we analyzed 4x4 pixel blocks of the images.

	<math>F_{after}=HF_{before}</math> 2		<math>F_{after}=HF_{before}</math> [2]

	We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.		We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.

	<math>D=\sum\|F_{before}(\omega)\|\omega-\sum\|HF_{before}(\omega)\|\omega</math> 3		<math>D=\sum\|F_{before}(\omega)\|\omega-\sum\|HF_{before}(\omega)\|\omega</math> [3]

	We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.		We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.

	~~<math>C(x,y) =c(x,y)^\beta s(x,y)^\gamma</math> 4~~		[[File:SSIM_2.png]]

	~~where~~:
	in which <math>\mu_a</math>, <math>\mu_b</math> and <math>\sigma_a</math>, <math>\sigma_b</math> are the means and standard deviations of the after-image region and the registered before-image region, respectively, and abis the covariance of these two image regions. ~~We used the standard constants when calculating SSIM (Wang 2003) [4~~]~~, where =1,=1,C2= (0.03)2, and C3=C2/2.~~

	===Implementation===		===Implementation===

imported>Psych2012: /* Photometric model */

2012-03-22T07:24:45Z

Photometric model

← Older revision		Revision as of 07:24, 22 March 2012
Line 63:		Line 63:

	where:		where:
	in which \mu_a, \mu_b and \sigma_a,\sigma_b are the means and standard deviations of the after-image region and the registered before-image region, respectively, and abis the covariance of these two image regions. We used the standard constants when calculating SSIM (Wang 2003) [4], where =1,=1,C2= (0.03)2, and C3=C2/2.		in which <math>\mu_a</math>, <math>\mu_b</math> and <math>\sigma_a</math>, <math>\sigma_b</math> are the means and standard deviations of the after-image region and the registered before-image region, respectively, and abis the covariance of these two image regions. We used the standard constants when calculating SSIM (Wang 2003) [4], where =1,=1,C2= (0.03)2, and C3=C2/2.

	===Implementation===		===Implementation===

imported>Psych2012: /* Photometric model */

2012-03-22T07:23:41Z

Photometric model

← Older revision		Revision as of 07:23, 22 March 2012
Line 63:		Line 63:

	where:		where:
	in which a, ~~banda~~,~~bare~~ the means and standard deviations of the after-image region and the registered before-image region, respectively, and abis the covariance of these two image regions. We used the standard constants when calculating SSIM (Wang 2003) [4], where =1,=1,C2= (0.03)2, and C3=C2/2.		in which \mu_a, \mu_b and \sigma_a,\sigma_b are the means and standard deviations of the after-image region and the registered before-image region, respectively, and abis the covariance of these two image regions. We used the standard constants when calculating SSIM (Wang 2003) [4], where =1,=1,C2= (0.03)2, and C3=C2/2.

	===Implementation===		===Implementation===

imported>Psych2012: /* Photometric model */

2012-03-22T07:22:34Z

Photometric model

← Older revision		Revision as of 07:22, 22 March 2012
Line 60:		Line 60:
	We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.		We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.

	<math>C(x,y) =c(x,y)s(x,y)</math> 4		<math>C(x,y) =c(x,y)^\beta s(x,y)^\gamma</math> 4

	where:		where:

imported>Psych2012: /* Photometric model */

2012-03-22T07:17:19Z

Photometric model

← Older revision		Revision as of 07:17, 22 March 2012
Line 56:		Line 56:
	We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.		We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.

	<math>D=\sum\|F_{before}\|-\sum\|HF_{before}\|</math> 3		<math>D=\sum\|F_{before}(\omega)\|\omega-\sum\|HF_{before}(\omega)\|\omega</math> 3

	We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.		We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.

imported>Psych2012: /* Data */

2012-03-22T07:15:41Z

Data

← Older revision		Revision as of 07:15, 22 March 2012
Line 196:		Line 196:

	==Data==		==Data==

			===Group Image Ratings===

	[http://white.stanford.edu/teach/index.php/File:group_ratings.zip User ratings recorded by group members for subsets of the images]		[http://white.stanford.edu/teach/index.php/File:group_ratings.zip User ratings recorded by group members for subsets of the images]

	~~[[File:Farid_ratings.zip]]: User ratings data graciously provided by Prof Farid. This set differs from that provided with the publication in that the numbering matches the author's photo sets.~~

	===Amazon Mechanical Turk observer ratings on retouching severity===		===Amazon Mechanical Turk observer ratings on retouching severity===
Line 204:		Line 205:

	Corresponding Masks: http://www.cs.dartmouth.edu/farid/downloads/publications/pnas11/masks.tar		Corresponding Masks: http://www.cs.dartmouth.edu/farid/downloads/publications/pnas11/masks.tar

			[[File:Farid_ratings.zip]]: User ratings data graciously provided by Prof Farid. This set differs from that provided with the publication in that the numbering matches the author's photo sets.

	Fifty observer ratings for each of the 468 before/after images used in Kee & Farid’s 2011 paper were acquired from the supplementary resources for the research paper. The authors gathered the user data using the process and observers described below:		Fifty observer ratings for each of the 468 before/after images used in Kee & Farid’s 2011 paper were acquired from the supplementary resources for the research paper. The authors gathered the user data using the process and observers described below:

imported>Psych2012: /* Photometric model */

2012-03-22T07:13:30Z

Photometric model

← Older revision		Revision as of 07:13, 22 March 2012
Line 56:		Line 56:
	We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.		We then determined the frequency response of each of the local filters H by using equation 3. If the result D is positive, the portion of the image corresponding to that H has been blurred, if it is negative, the local part of the image has been sharpened.

	<math>D=\|F_{before}\|-\|HF_{before}\|</math> 3		<math>D=\sum\|F_{before}\|-\sum\|HF_{before}\|</math> 3

	We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.		We also used a generic measure of local image similarity (SSIM) to detect photometric modifications not captured by equation 3. The SSIM measure embodies contrast and structural modifications as follows in equation 4.