AAMToolbox statistical model generator - Revision history

AndrewBangham at 16:03, 13 February 2012

2012-02-13T16:03:20Z

← Older revision		Revision as of 16:03, 13 February 2012
Line 1:		Line 1:
	[[Tutorials on the Shape modelling toolbox#1\|Back to Tutorial pages]]<br>		[[Tutorials on the Shape modelling toolbox#1\|Back to Tutorial pages]]<br>
	=Statistical shape generator (PCA)=		=<span style="color:Navy;">Statistical shape generator (PCA)</span>=
	==Principle component analysis of point-models and image pixel values (appearance models)==		==Principle component analysis of point-models and image pixel values (appearance models)==
	To run the PCA select '''Stats Model Generator''' from the AAMToolbox control panel: <br><br>		To run the PCA select '''Stats Model Generator''' from the AAMToolbox control panel: <br><br>
	=Inputs and outputs=		=<span style="color:Navy;">Inputs and outputs</span>=
	*Shape model		*Shape model
	**Input: landmark points, i.e. the point-models for all images of interest.		**Input: landmark points, i.e. the point-models for all images of interest.

AndrewBangham at 16:01, 13 February 2012

2012-02-13T16:01:38Z

AndrewBangham at 15:57, 13 February 2012

2012-02-13T15:57:34Z

← Older revision		Revision as of 15:57, 13 February 2012
Line 12:		Line 12:
	=Statistical Model Generator control panel.=		=Statistical Model Generator control panel.=
	Showing the steps (in red) to build a statistical model.<br>		Showing the steps (in red) to build a statistical model.<br>
			Notice: the set of landmarks represented as stars will be included in the shape model, in this case all of them. For this model you can accept the default settings and jump directly to step 3 and select all the images. Then compute the principle component analysis (PCA) using step 4 and return to the AAMToolbox on the way to viewing the results.<br><br>
	[[File:Model_generator_AAMToolbox.png\|822px\|The Model generator control panel]]<br><br>		[[File:Model_generator_AAMToolbox.png\|822px\|The Model generator control panel]]<br><br>
	==Steps to build a shape model using principle component analysis (PCA)==		==Steps to build a shape model using principle component analysis (PCA)==

AndrewBangham at 13:31, 13 February 2012

2012-02-13T13:31:14Z

← Older revision		Revision as of 13:31, 13 February 2012
Line 26:		Line 26:
	**If modelling a subset of landmark points it can be useful to place the model at the mean position of all the landmarks. This can be selected by ticking 'include all points in mean'. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)<br>		**If modelling a subset of landmark points it can be useful to place the model at the mean position of all the landmarks. This can be selected by ticking 'include all points in mean'. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)<br>
	===Step 3) Point models and images contributing to the model===		===Step 3) Point models and images contributing to the model===
	*'''Select the image point models''' you want to include. Usually all of them.		*'''Select the image point models''' you want to include. Usually all of them. Here, select all.
	[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>		[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>
	**You will invited to give a name to the model you are about to compute		**You will invited to give a name to the model you are about to compute. You choose a name.
	===Step 4) Generate Model===		===Step 4) Generate Model===
	===Step 5) Exit===		===Step 5) Exit===
	*The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''		*The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''

AndrewBangham at 13:30, 13 February 2012

2012-02-13T13:30:19Z

← Older revision		Revision as of 13:30, 13 February 2012
Line 24:		Line 24:
	***Select the number of pixels to be used when forming the mean image intensity (appearance).		***Select the number of pixels to be used when forming the mean image intensity (appearance).
	**'''Point set'''. The AAMToolbox supports the concept of point model 'sets'. In other words, subsets of landmarks can be grouped together into a set. Set_1 is always all the points. The points that are currently selected are shown as stars. Here, all the points are selected. The options are discussed in the Tutorial on 'sets'.		**'''Point set'''. The AAMToolbox supports the concept of point model 'sets'. In other words, subsets of landmarks can be grouped together into a set. Set_1 is always all the points. The points that are currently selected are shown as stars. Here, all the points are selected. The options are discussed in the Tutorial on 'sets'.
	**If modelling a subset of landmark points it can be useful to place the model at the mean position of all the landmarks. This can be selected by ticking 'include all points in mean'. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)		**If modelling a subset of landmark points it can be useful to place the model at the mean position of all the landmarks. This can be selected by ticking 'include all points in mean'. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)<br>
	===Step 3) Point models and images ~~to contribute~~ to the model		===Step 3) Point models and images contributing to the model===
	*'''Select the image point models''' you want to include. Usually all of them.		*'''Select the image point models''' you want to include. Usually all of them.
	[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>		[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>

AndrewBangham at 13:28, 13 February 2012

2012-02-13T13:28:51Z

← Older revision		Revision as of 13:28, 13 February 2012
Line 19:		Line 19:
	For the '''cartoon faces we normalise for all three'''.		For the '''cartoon faces we normalise for all three'''.
	===Step 2) Type of model===		===Step 2) Type of model===
	In this case the statistical '''Shape''' model will be generated from the point models alone. In other cases, we might want to include the image colours, pixel by pixel, in the PCA in which case we create two models '''Shape and Appearance''' (AAM).		In this case the statistical '''Shape''' model will be generated from the point models alone. In other cases, we might want to include the image colours, pixel by pixel, in the PCA in which case we create two models '''Shape and Appearance''' (AAM) and accept the defaults for all the other options. However, ...
	**If AAM then		**If AAM then
	***Select whether to mask out the image not within the a boundary set by the outermost landmark points		***Select whether to mask out the image not within the a boundary set by the outermost landmark points

AndrewBangham at 13:26, 13 February 2012

2012-02-13T13:26:50Z

← Older revision		Revision as of 13:26, 13 February 2012
Line 2:		Line 2:
	=Statistical shape generator (PCA)=		=Statistical shape generator (PCA)=
	==Principle component analysis of point-models and image pixel values (appearance models)==		==Principle component analysis of point-models and image pixel values (appearance models)==
	~~From~~ the ~~AAMToolbox workflow control panel elect~~ '''Stats Model Generator''': <br><br>		To run the PCA select '''Stats Model Generator''' from the AAMToolbox control panel: <br><br>
	=Inputs and outputs=		=Inputs and outputs=
	*Shape model		*Shape model

AndrewBangham at 13:25, 13 February 2012

2012-02-13T13:25:56Z

← Older revision		Revision as of 13:25, 13 February 2012
Line 1:		Line 1:
	[[Tutorials on the Shape modelling toolbox#1\|Back to Tutorial pages]]<br>		[[Tutorials on the Shape modelling toolbox#1\|Back to Tutorial pages]]<br>
	=Statistical shape generator (PCA)=		=Statistical shape generator (PCA)=
	==~~Generate a new PCA model from the AAMToolbox control panel~~==		==Principle component analysis of point-models and image pixel values (appearance models)==
	From the AAMToolbox workflow control panel elect '''Stats Model Generator''': <br><br>		From the AAMToolbox workflow control panel elect '''Stats Model Generator''': <br><br>
			=Inputs and outputs=
			*Shape model
			**Input: landmark points, i.e. the point-models for all images of interest.
			**Output: mean shape and associated shape model that captures deviations from the mean. Enough principle components are included to account for 95% of the variance from the mean (set using Options).
			*Appearance model
			**Input: landmark points, i.e. the point-models together with the image intensity within triangles determined by a Delauny triangulation of the points.
			**Output: mean appearance and associated appearance model that captures deviations from the mean.
	=Statistical Model Generator control panel.=		=Statistical Model Generator control panel.=
	Showing the steps (in red) to build a statistical model.<br>		Showing the steps (in red) to build a statistical model.<br>

AndrewBangham at 12:40, 13 February 2012

2012-02-13T12:40:59Z

← Older revision		Revision as of 12:40, 13 February 2012
Line 10:		Line 10:
	===Step 1) Options===		===Step 1) Options===
	[[File:Stats_Model_Options_AAMToolbox.png\|500px\|The Model generator control panel]]<br><br>		[[File:Stats_Model_Options_AAMToolbox.png\|500px\|The Model generator control panel]]<br><br>
	*~~Select~~ the ~~required normalisation using Procrustes method~~. ~~Normally,~~ the ~~point model positions in~~ the ~~image (sideways and up-down) are not important~~ and we '''~~normalise~~ the ~~translation~~. ~~Similarly~~ the ~~rotation and scale~~'''. ~~However~~, ~~sometimes~~ the ~~scale is an important part~~ of the ~~model - for example, when analysing growing leaves. In this case scale is critical and~~ should be ~~unticked~~.<br>		*The first step in PCA is to align all the point models. In this case we assume that the images should be normalised to the''' same scale, same rotation and same translational''' position. This might not always be the case. If we want to model leaf growth then the point model should '''not''' be scaled. Alternatively, if we want to capture the pose of a face portrait the point models should not be rotationally aligned.<br>
			For the '''cartoon faces we normalise for all three'''.
	===Step 2) Type of model===		===Step 2) Type of model===
	*Select either a shape model or shape model and ~~a separate appearance model~~ (AAM).		In this case the statistical '''Shape''' model will be generated from the point models alone. In other cases, we might want to include the image colours, pixel by pixel, in the PCA in which case we create two models '''Shape and Appearance''' (AAM).
	**If AAM then		**If AAM then
	***Select whether to mask out the image not within the a boundary set by the outermost landmark points		***Select whether to mask out the image not within the a boundary set by the outermost landmark points
	***Select the number of pixels to be used when forming the mean image intensity (appearance).		***Select the number of pixels to be used when forming the mean image intensity (appearance).
	**If modelling a subset of landmark points ~~choose whether~~ to ~~find~~ the mean position of the ~~point models using~~ all ~~the landmark~~ points ~~(use~~ mean ~~positions for points not in Set) or whether to ignore landmarks not in the subset~~. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)		**'''Point set'''. The AAMToolbox supports the concept of point model 'sets'. In other words, subsets of landmarks can be grouped together into a set. Set_1 is always all the points. The points that are currently selected are shown as stars. Here, all the points are selected. The options are discussed in the Tutorial on 'sets'.
			**If modelling a subset of landmark points it can be useful to place the model at the mean position of all the landmarks. This can be selected by ticking 'include all points in mean'. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)
	===Step 3) Point models and images to contribute to the model		===Step 3) Point models and images to contribute to the model
	*'''Select the image point models''' you want to include. Usually all of them.		*'''Select the image point models''' you want to include. Usually all of them.
Line 24:		Line 26:
	===Step 5) Exit===		===Step 5) Exit===
	*The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''		*The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''

	~~===Procrustes===~~
	The first step in PCA is to align all the point models. In this case we assume that the images should be normalised to the''' same scale, same rotation and same translational''' position. This might not always be the case. If we want to model leaf growth then the point model should '''not''' be scaled. Alternatively, if we want to capture the pose of a face portrait the point models should not be rotationally aligned.
	~~===Model Selection===~~
	In this case the statistical '''Shape''' model will be generated from the point models alone. In other cases, we might want to include the image colours, pixel by pixel, in the PCA in which case we model '''Shape and Appearance'''.
	~~===Options===~~
	~~Verbose is useful for large Shape and Appearance models because it allows us to monitor progress. The other options control the way points are divided into sets.<br>~~
	'''Point set'''. The AAMToolbox supports the concept of point model 'sets'. In other words, selections of points can be grouped together into a set. Set_1 is always all the points. The points that are currently selected are shown as stars. Here, all the points are selected. The options are discussed in the Tutorial on 'sets'.

AndrewBangham at 11:13, 13 February 2012

2012-02-13T11:13:14Z

← Older revision		Revision as of 11:13, 13 February 2012
Line 11:		Line 11:
	[[File:Stats_Model_Options_AAMToolbox.png\|500px\|The Model generator control panel]]<br><br>		[[File:Stats_Model_Options_AAMToolbox.png\|500px\|The Model generator control panel]]<br><br>
	*Select the required normalisation using Procrustes method. Normally, the point model positions in the image (sideways and up-down) are not important and we '''normalise the translation. Similarly the rotation and scale'''. However, sometimes the scale is an important part of the model - for example, when analysing growing leaves. In this case scale is critical and should be unticked.<br>		*Select the required normalisation using Procrustes method. Normally, the point model positions in the image (sideways and up-down) are not important and we '''normalise the translation. Similarly the rotation and scale'''. However, sometimes the scale is an important part of the model - for example, when analysing growing leaves. In this case scale is critical and should be unticked.<br>
			===Step 2) Type of model===
	*Select either a shape model or shape model and a separate appearance model (AAM).		*Select either a shape model or shape model and a separate appearance model (AAM).
	**If AAM then		**If AAM then
Line 16:		Line 17:
	***Select the number of pixels to be used when forming the mean image intensity (appearance).		***Select the number of pixels to be used when forming the mean image intensity (appearance).
	**If modelling a subset of landmark points choose whether to find the mean position of the point models using all the landmark points (use mean positions for points not in Set) or whether to ignore landmarks not in the subset. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)		**If modelling a subset of landmark points choose whether to find the mean position of the point models using all the landmark points (use mean positions for points not in Set) or whether to ignore landmarks not in the subset. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)
	#'''Select the image point models''' you want to include. Usually all of them.		===Step 3) Point models and images to contribute to the model
			*'''Select the image point models''' you want to include. Usually all of them.
	[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>		[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>
			**You will invited to give a name to the model you are about to compute
	#*You will invited to give a name to the model you are about to compute		===Step 4) Generate Model===
	~~#Click '''~~Generate Model~~'''~~		===Step 5) Exit===
	~~#'''Close''' the box to save the model and return to the AAMToolbox~~		*The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''
	#The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''

	===Procrustes===		===Procrustes===

← Older revision		Revision as of 16:01, 13 February 2012
Line 10:		Line 10:
	**Input: landmark points, i.e. the point-models together with the image intensity within triangles determined by a Delauny triangulation of the points.		**Input: landmark points, i.e. the point-models together with the image intensity within triangles determined by a Delauny triangulation of the points.
	**Output: mean appearance and associated appearance model that captures deviations from the mean.		**Output: mean appearance and associated appearance model that captures deviations from the mean.
	=Statistical Model Generator control panel.=		=<span style="color:Navy;">Statistical Model Generator control panel.</span>=
	Showing the steps (in red) to build a statistical model.<br>		Showing the steps (in red) to build a statistical model.<br>
	Notice: the set of landmarks represented as stars will be included in the shape model, in this case all of them. For this model you can accept the default settings and jump directly to step 3 and select all the images. Then compute the principle component analysis (PCA) using step 4 and return to the AAMToolbox on the way to viewing the results.<br><br>		Notice: the set of landmarks represented as stars will be included in the shape model, in this case all of them. For this model you can accept the default settings and jump directly to step 3 and select all the images. Then compute the principle component analysis (PCA) using step 4 and return to the AAMToolbox on the way to viewing the results.<br><br>
	[[File:Model_generator_AAMToolbox.png\|822px\|The Model generator control panel]]<br><br>		[[File:Model_generator_AAMToolbox.png\|822px\|The Model generator control panel]]<br><br>
	==Steps to build a shape model using principle component analysis (PCA)==		==Steps to build a shape model using principle component analysis (PCA)==
	===Step 1) Options===		===<span style="color:FireBrick;">Step 1) Options</span>===
	[[File:Stats_Model_Options_AAMToolbox.png\|500px\|The Model generator control panel]]<br><br>		[[File:Stats_Model_Options_AAMToolbox.png\|500px\|The Model generator control panel]]<br><br>
	*The first step in PCA is to align all the point models. In this case we assume that the images should be normalised to the''' same scale, same rotation and same translational''' position. This might not always be the case. If we want to model leaf growth then the point model should '''not''' be scaled. Alternatively, if we want to capture the pose of a face portrait the point models should not be rotationally aligned.<br>		*The first step in PCA is to align all the point models. In this case we assume that the images should be normalised to the''' same scale, same rotation and same translational''' position. This might not always be the case. If we want to model leaf growth then the point model should '''not''' be scaled. Alternatively, if we want to capture the pose of a face portrait the point models should not be rotationally aligned.<br>
	For the '''cartoon faces we normalise for all three'''.		For the '''cartoon faces we normalise for all three'''.
	===Step 2) Type of model===		===<span style="color:FireBrick;">Step 2) Type of model</span>===
	In this case the statistical '''Shape''' model will be generated from the point models alone. In other cases, we might want to include the image colours, pixel by pixel, in the PCA in which case we create two models '''Shape and Appearance''' (AAM) and accept the defaults for all the other options. However, ...		In this case the statistical '''Shape''' model will be generated from the point models alone. In other cases, we might want to include the image colours, pixel by pixel, in the PCA in which case we create two models '''Shape and Appearance''' (AAM) and accept the defaults for all the other options. However, ...
	**If AAM then		**If AAM then
Line 26:		Line 26:
	**'''Point set'''. The AAMToolbox supports the concept of point model 'sets'. In other words, subsets of landmarks can be grouped together into a set. Set_1 is always all the points. The points that are currently selected are shown as stars. Here, all the points are selected. The options are discussed in the Tutorial on 'sets'.		**'''Point set'''. The AAMToolbox supports the concept of point model 'sets'. In other words, subsets of landmarks can be grouped together into a set. Set_1 is always all the points. The points that are currently selected are shown as stars. Here, all the points are selected. The options are discussed in the Tutorial on 'sets'.
	**If modelling a subset of landmark points it can be useful to place the model at the mean position of all the landmarks. This can be selected by ticking 'include all points in mean'. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)<br>		**If modelling a subset of landmark points it can be useful to place the model at the mean position of all the landmarks. This can be selected by ticking 'include all points in mean'. (A subset of landmark points is selected in the AAMToolbox:Point Model Editor.)<br>
	===Step 3) Point models and images contributing to the model===		===<span style="color:FireBrick;">Step 3) Point models and images contributing to the model</span>===
	*'''Select the image point models''' you want to include. Usually all of them. Here, select all.		*'''Select the image point models''' you want to include. Usually all of them. Here, select all.
	[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>		[[File:PM_selection_Model_generator_AAMToolbox.png\|300px\|The Model generator control panel]]<br><br>
	**You will invited to give a name to the model you are about to compute. You choose a name.		**You will invited to give a name to the model you are about to compute. You choose a name.
	===Step 4) Generate Model===		===<span style="color:FireBrick;">Step 4) Generate Model</span>===
	===Step 5) Exit===		===<span style="color:FireBrick;">Step 5) Exit</span>===
	*The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''		*The next step is to view results from the model, AAMToolbox: '''View Stats Mode'''