BoofCV - User contributions [en]

Contributing

2012-04-22T15:26:10Z

Maarten Van Lier: /* Specific List */

= Ways to Contribute =
BoofCV is an open source project and in order for it to grow its users need contribute to the project! Even if you don't know computer vision or how to program there are many ways you can contribute.

== Specific List ==

The follow tasks are fairly specific and can be done without requiring major changes to the library:

# Write about BoofCV!
## BoofCV is very new and not well know, so post about it on your blog/twitter/bathroom wall/tattoo
## If you use BoofCV in your project link back to this webpage to improve its page rank on search engines
# [[Android_support|Android support]]
## Identify incompatibilities with Android's API
## Help come up with solutions
## Create Android examples
# Improve Wiki and Documentation
## Create a wiki account by clicking on the left
## After the account has been approved edit away
## Approval is required to counter act all the spam bots
# Wavelets
## Arbitrary length Coiflet coefficients
## Arbitrary length Daub coefficients
## Verify that Biorthogonal Daub is buggy and fix
## Don't add any addition jar file requires for wavelet tasks
# Corner detectors
## Add option to process image border in SSD feature detectors
# Modify BRIEF so that it can describe feature partially outside of image
# Add fast non-maximum suppression algorithm
# See in code comments for improving Canny edge's speed

== Non-Coding Tasks ==

Non-coding tasks can be done by just about anyone. Typically these involve updating documentation and reporting problems. Design changes should be suggested to the message board.

# Reporting bugs as they are found.
# Fixing spelling and grammar errors on the wiki.
# Checking links on the wiki.
# Adding tutorials to the wiki.
# Quality control on JavaDOC.
# Suggesting ways to improve the API.

== Coding Tasks ==

There is plenty of room for improvement left in BoofCV. If you which to donate code or tweak the existing code you should sign up on the BoofCV message board (see the [[main page]] for a link) and join the discussion. An important part of BoofCV is validating the correctness of the library. Making sure that each piece of code does what it says it does or that it is a correct implementation of the algorithm cited requires time and many eyes.

# Fixing bugs which have been reported.
# Adding unit tests.
# Adding system level tests.
# Implementing new algorithms.
# Validating the correctness of algorithms and implementations.
# Improve demo [[List of Applets|Applets]].

Download

2012-04-22T15:25:19Z

Maarten Van Lier: /* Latest Release */

= Latest Release =

BoofCV is available through source code from its central repository or as formal releases:

Download Latest Release:
* [http://sourceforge.net/projects/boofcv/files/v0.6/boofcv-v0.6-libs.zip/download Compiled Jar and dependencies]
* [http://sourceforge.net/projects/boofcv/files/v0.6/boofcv-v0.6-docs.zip/download JavaDOC]
* [http://sourceforge.net/projects/boofcv/files/v0.6/boofcv-v0.6-src.zip/download Source Code]
* [[BoofCV_Change_Log|Change Log]]

See below for how to checkout code from the git repository.

= Bleeding Edge =

BoofCV uses github for its central repository. Before you run to github and clone the repository, don't forget to checkout the data submodule! Those data files are needed to run the example code.

git repository: https://github.com/lessthanoptimal/BoofCV

To checkout the latest source code type the following:
<pre>
git clone git://github.com/lessthanoptimal/BoofCV.git boofcv
cd boofcv
git submodule init
git submodule update
</pre>

Larger data files have been put into a separate git repository and the submodule related lines are required to download them. Those data files
will appear inside the BoofCV/data directory.

External jars which BoofCV relies upon are not always updated in the repository and can cause problems when trying to compile the code. It is recommended that [https://github.com/lessthanoptimal/GeoRegression GeoRegression] also be checked out from its repository if you are working with the latest BoofCV repository code.

= Past Releases =

{| border="1" width=600px cellpadding="4"
! Version !! Link !! Description
|-
| Alpha v0.6 || [https://sourceforge.net/projects/boofcv/files/v0.6 Link] || Camera calibration, 3D Vision, Performance, and more
|-
| Alpha v0.5 || [https://sourceforge.net/projects/boofcv/files/v0.5 Link] || Improvements to SURF and non-linear optimization
|-
| Alpha v0.4 || [https://sourceforge.net/projects/boofcv/files/v0.4 Link] || Fixed bugs in SURF and MultiSpectral images
|-
| Alpha v0.3 || [https://sourceforge.net/projects/boofcv/files/v0.3 Link] || More color and camera calibration support
|-
| Alpha v0.2 || [https://sourceforge.net/projects/boofcv/files/v0.2 Link] || API improvements and geometric vision
|-
| Alpha v0.1 || [https://sourceforge.net/projects/boofcv/files/v0.1 Link] || First formal publish release
|}

Download

2012-04-22T15:24:53Z

Maarten Van Lier: /* Latest Release */

= Latest Release =

BoofCV is available through source code from its central repository or as formal releases:

Download Latest Release:
* [http://sourceforge.net/projects/boofcv/files/v0.6/boofcv-v0.6-libs.zip/download Compiled Jar and dependencies]
* [http://sourceforge.net/projects/boofcv/files/v0.6/boofcv-v0.6-docs.zip/download JavaDOC]
* [http://sourceforge.net/projects/boofcv/files/v0.6/boofcv-v0.6-src.zip/download Source Code]
* [[BoofCV_Change_Log|Change Log]]

See below for how to checkout code from the git repository.

Android support has its own page: [[Android_support|Android support]]

= Bleeding Edge =

BoofCV uses github for its central repository. Before you run to github and clone the repository, don't forget to checkout the data submodule! Those data files are needed to run the example code.

git repository: https://github.com/lessthanoptimal/BoofCV

To checkout the latest source code type the following:
<pre>
git clone git://github.com/lessthanoptimal/BoofCV.git boofcv
cd boofcv
git submodule init
git submodule update
</pre>

Larger data files have been put into a separate git repository and the submodule related lines are required to download them. Those data files
will appear inside the BoofCV/data directory.

External jars which BoofCV relies upon are not always updated in the repository and can cause problems when trying to compile the code. It is recommended that [https://github.com/lessthanoptimal/GeoRegression GeoRegression] also be checked out from its repository if you are working with the latest BoofCV repository code.

= Past Releases =

{| border="1" width=600px cellpadding="4"
! Version !! Link !! Description
|-
| Alpha v0.6 || [https://sourceforge.net/projects/boofcv/files/v0.6 Link] || Camera calibration, 3D Vision, Performance, and more
|-
| Alpha v0.5 || [https://sourceforge.net/projects/boofcv/files/v0.5 Link] || Improvements to SURF and non-linear optimization
|-
| Alpha v0.4 || [https://sourceforge.net/projects/boofcv/files/v0.4 Link] || Fixed bugs in SURF and MultiSpectral images
|-
| Alpha v0.3 || [https://sourceforge.net/projects/boofcv/files/v0.3 Link] || More color and camera calibration support
|-
| Alpha v0.2 || [https://sourceforge.net/projects/boofcv/files/v0.2 Link] || API improvements and geometric vision
|-
| Alpha v0.1 || [https://sourceforge.net/projects/boofcv/files/v0.1 Link] || First formal publish release
|}

Android support

2012-04-22T15:22:47Z

Maarten Van Lier: /* Partial Android Port of BoofCV v0.2 */

= Android Support =

This page is a summary of what needs to be done to port BoofCV to Android.
The required libraries (EJML, GeoRegression and libpja) can be used without modification on Android.
BoofCV itself is mostly compatible (all algorithms can just be used), but needs some modifications to be able to run:
* Since there is no java swing library, all examples have to be adapted to use the Android activity system.
* Since Android uses Bitmaps instead of BufferedImage, the code that converts BufferedImage objects to the internal representation needs to be changed to work with Bitmap objects.

== Classes that need change ==
Please note that this list is based on the experience of a partial port of BoofCV v0.2.
* core.image
** ConvertBufferedImage needs to use Bitmaps
* gui
** This whole package needs to use Android Activities to show the results instead of java swing windows
* io
** io.image
*** UtilIO needs to use an activity for selectFile, so a file can be selected from the Android file system
*** ImageListManager needs to use Bitmap instead of BufferedImage
*** PlaybackImageSequance needs an activity to play back the sequence
*** ProcessImageSequence needs an activity to process and show the sequence
*** SelectInputImageToolBar is very swing specific and maybe not even needed for Android
*** SimpleImageSequence needs to use Bitmap instead of BufferedImage
*** UtilImageIO needs to use Bitmap instead of BufferedImage
** io.video
*** BoofVideoManager => ?
*** VideoInterface => ?
*** VideoListManager => ?
** io.wrapper.images
*** BufferedFileImageSequence probably just needs Bitmap instead of BufferedImage
*** JpegByteImageSequence probably just needs Bitmap instead of BufferedImage
*** LoadFileImageSequance probably just needs Bitmap instead of BufferedImage
* testing
** BoofTesting needs Bitmap instead of BufferedImage

== Partial Android Port of BoofCV v0.2 ==
For his [[Projects_Using_BoofCV|master thesis]], [[User:Maarten_Van_Lier|Maarten]] made a partial port of BoofCV v0.2 to Android.
In this port, the following was done to get it working:
* Just use the EJML, GeoRegression and libpja libraries that BoofCV uses, they work on Android as is.
* Modify the core.image.ConvertBufferedImage class to the core.image.ConvertBitmap class. This new class uses BufferedImages and Android API calls to do the same as the ConvertBufferedImage class. The implementation is not as efficient as the original one, though.
* Leave out the gui package because of the java swing windows that were used. It was not needed for the thesis.
* Leave out problematic methods in io.image.UtilIO and remove other classes in io.image, io.video and io.wrapper.images, because they use java swing window calls. They were not needed for the thesis.
* Leave out the methods in testing.BoofTesting that used the BufferedImage class.

Basically, code for loading images was ported, while code to show windows was left out.
The code is available here[http://dl.dropbox.com/u/16413201/BoofCV%20v0.2%20Partial%20Android%20Port.zip]. Note that this is an incomplete port that may not have what you need, or may not work as expected.

Android support

2012-04-22T15:21:54Z

Maarten Van Lier: /* Partial Android Port of BoofCV v0.2 */

= Android Support =

This page is a summary of what needs to be done to port BoofCV to Android.
The required libraries (EJML, GeoRegression and libpja) can be used without modification on Android.
BoofCV itself is mostly compatible (all algorithms can just be used), but needs some modifications to be able to run:
* Since there is no java swing library, all examples have to be adapted to use the Android activity system.
* Since Android uses Bitmaps instead of BufferedImage, the code that converts BufferedImage objects to the internal representation needs to be changed to work with Bitmap objects.

== Classes that need change ==
Please note that this list is based on the experience of a partial port of BoofCV v0.2.
* core.image
** ConvertBufferedImage needs to use Bitmaps
* gui
** This whole package needs to use Android Activities to show the results instead of java swing windows
* io
** io.image
*** UtilIO needs to use an activity for selectFile, so a file can be selected from the Android file system
*** ImageListManager needs to use Bitmap instead of BufferedImage
*** PlaybackImageSequance needs an activity to play back the sequence
*** ProcessImageSequence needs an activity to process and show the sequence
*** SelectInputImageToolBar is very swing specific and maybe not even needed for Android
*** SimpleImageSequence needs to use Bitmap instead of BufferedImage
*** UtilImageIO needs to use Bitmap instead of BufferedImage
** io.video
*** BoofVideoManager => ?
*** VideoInterface => ?
*** VideoListManager => ?
** io.wrapper.images
*** BufferedFileImageSequence probably just needs Bitmap instead of BufferedImage
*** JpegByteImageSequence probably just needs Bitmap instead of BufferedImage
*** LoadFileImageSequance probably just needs Bitmap instead of BufferedImage
* testing
** BoofTesting needs Bitmap instead of BufferedImage

== Partial Android Port of BoofCV v0.2 ==
For his [[Projects_Using_BoofCV|master thesis]], [[User:Maarten_Van_Lier|Maarten]] made a partial port of BoofCV v0.2 to Android.
In this port, the following was done to get it working:
* Just use the EJML, GeoRegression and libpja libraries that BoofCV uses, they work on Android as is.
* Modify the core.image.ConvertBufferedImage class to the core.image.ConvertBitmap class. This new class uses BufferedImages and Android API calls to do the same as the ConvertBufferedImage class. The implementation is not as efficient as the original one, though.
* Leave out the gui package because of the java swing windows that were used. It was not needed for the thesis.
* Leave out problematic methods in io.image.UtilIO and remove other classes in io.image, io.video and io.wrapper.images, because they use java swing window calls. They were not needed for the thesis.
* Leave out the methods in testing.BoofTesting that used the BufferedImage class.

Basically, code for loading images was ported, while code to show windows was left out.
The code is available [http://dl.dropbox.com/u/16413201/BoofCV%20v0.2%20Partial%20Android%20Port.zip|here here]. Note that this is an incomplete port that may not have what you need, or may not work as expected.

Android support

2012-04-22T15:15:34Z

Maarten Van Lier: /* Android Support */

= Android Support =

This page is a summary of what needs to be done to port BoofCV to Android.
The required libraries (EJML, GeoRegression and libpja) can be used without modification on Android.
BoofCV itself is mostly compatible (all algorithms can just be used), but needs some modifications to be able to run:
* Since there is no java swing library, all examples have to be adapted to use the Android activity system.
* Since Android uses Bitmaps instead of BufferedImage, the code that converts BufferedImage objects to the internal representation needs to be changed to work with Bitmap objects.

== Classes that need change ==
Please note that this list is based on the experience of a partial port of BoofCV v0.2.
* core.image
** ConvertBufferedImage needs to use Bitmaps
* gui
** This whole package needs to use Android Activities to show the results instead of java swing windows
* io
** io.image
*** UtilIO needs to use an activity for selectFile, so a file can be selected from the Android file system
*** ImageListManager needs to use Bitmap instead of BufferedImage
*** PlaybackImageSequance needs an activity to play back the sequence
*** ProcessImageSequence needs an activity to process and show the sequence
*** SelectInputImageToolBar is very swing specific and maybe not even needed for Android
*** SimpleImageSequence needs to use Bitmap instead of BufferedImage
*** UtilImageIO needs to use Bitmap instead of BufferedImage
** io.video
*** BoofVideoManager => ?
*** VideoInterface => ?
*** VideoListManager => ?
** io.wrapper.images
*** BufferedFileImageSequence probably just needs Bitmap instead of BufferedImage
*** JpegByteImageSequence probably just needs Bitmap instead of BufferedImage
*** LoadFileImageSequance probably just needs Bitmap instead of BufferedImage
* testing
** BoofTesting needs Bitmap instead of BufferedImage

== Partial Android Port of BoofCV v0.2 ==
For his [[Projects_Using_BoofCV|master thesis]], [[User:Maarten_Van_Lier|Maarten]] made a partial port of BoofCV v0.2 to Android.
In this port, the following was done to get it working:
* Just use the EJML, GeoRegression and libpja libraries that BoofCV uses, they work on Android as is.
* Modify the core.image.ConvertBufferedImage class to the core.image.ConvertBitmap class. This new class uses BufferedImages and Android API calls to do the same as the ConvertBufferedImage class. The implementation is not as efficient as the original one, though.
* Leave out the gui package because of the java swing windows that were used. It was not needed for the thesis.
* Leave out problematic methods in io.image.UtilIO and remove other classes in io.image, io.video and io.wrapper.images, because they use java swing window calls. They were not needed for the thesis.
* Leave out the methods in testing.BoofTesting that used the BufferedImage class.

Basically, code for loading images was ported, while code to show windows was left out.
The code is available here (TODO: add link). Note that this is an incomplete port that may not have what you need, or may not work as expected.

Android support

2012-04-22T15:14:15Z

Maarten Van Lier: /* Android Support */

= Android Support =

This page is a summary of what needs to be done to port BoofCV to Android.
The required libraries (EJML, GeoRegression and libpja) can be used without modification on Android.
BoofCV itself is mostly compatible (all algorithms can just be used), but needs some modifications to be able to run:
* Since there is no java swing library, all examples have to be adapted to use the Android activity system.
* Since Android uses Bitmaps instead of BufferedImage, the code that converts BufferedImage objects to the internal representation needs to be changed to work with Bitmap objects.

== Classes that need change ==
* core.image
** ConvertBufferedImage needs to use Bitmaps
* gui
** This whole package needs to use Android Activities to show the results instead of java swing windows
* io
** io.image
*** UtilIO needs to use an activity for selectFile, so a file can be selected from the Android file system
*** ImageListManager needs to use Bitmap instead of BufferedImage
*** PlaybackImageSequance needs an activity to play back the sequence
*** ProcessImageSequence needs an activity to process and show the sequence
*** SelectInputImageToolBar is very swing specific and maybe not even needed for Android
*** SimpleImageSequence needs to use Bitmap instead of BufferedImage
*** UtilImageIO needs to use Bitmap instead of BufferedImage
** io.video
*** BoofVideoManager => ?
*** VideoInterface => ?
*** VideoListManager => ?
** io.wrapper.images
*** BufferedFileImageSequence probably just needs Bitmap instead of BufferedImage
*** JpegByteImageSequence probably just needs Bitmap instead of BufferedImage
*** LoadFileImageSequance probably just needs Bitmap instead of BufferedImage
* testing
** BoofTesting needs Bitmap instead of BufferedImage

== Partial Android Port of BoofCV v0.2 ==
For his [[Projects_Using_BoofCV|master thesis]], [[User:Maarten_Van_Lier|Maarten]] made a partial port of BoofCV v0.2 to Android.
In this port, the following was done to get it working:
* Just use the EJML, GeoRegression and libpja libraries that BoofCV uses, they work on Android as is.
* Modify the core.image.ConvertBufferedImage class to the core.image.ConvertBitmap class. This new class uses BufferedImages and Android API calls to do the same as the ConvertBufferedImage class. The implementation is not as efficient as the original one, though.
* Leave out the gui package because of the java swing windows that were used. It was not needed for the thesis.
* Leave out problematic methods in io.image.UtilIO and remove other classes in io.image, io.video and io.wrapper.images, because they use java swing window calls. They were not needed for the thesis.
* Leave out the methods in testing.BoofTesting that used the BufferedImage class.

Basically, code for loading images was ported, while code to show windows was left out.
The code is available here (TODO: add link). Note that this is an incomplete port that may not have what you need, or may not work as expected.

Android support

2012-04-22T15:13:19Z

Maarten Van Lier: Initial version of the Android port page.

= Android Support =

This page is a summary of what needs to be done to port BoofCV to Android.
The required libraries (EJML, GeoRegression and libpja) can be used without modification on Android.
BoofCV itself needs some modifications to be able to run:
* Since there is no java swing library, all examples have to be adapted to use the Android activity system.
* Since Android uses Bitmaps instead of BufferedImage, the code that converts BufferedImage objects to the internal representation needs to be changed to work with Bitmap objects.

== Classes that need change ==
* core.image
** ConvertBufferedImage needs to use Bitmaps
* gui
** This whole package needs to use Android Activities to show the results instead of java swing windows
* io
** io.image
*** UtilIO needs to use an activity for selectFile, so a file can be selected from the Android file system
*** ImageListManager needs to use Bitmap instead of BufferedImage
*** PlaybackImageSequance needs an activity to play back the sequence
*** ProcessImageSequence needs an activity to process and show the sequence
*** SelectInputImageToolBar is very swing specific and maybe not even needed for Android
*** SimpleImageSequence needs to use Bitmap instead of BufferedImage
*** UtilImageIO needs to use Bitmap instead of BufferedImage
** io.video
*** BoofVideoManager => ?
*** VideoInterface => ?
*** VideoListManager => ?
** io.wrapper.images
*** BufferedFileImageSequence probably just needs Bitmap instead of BufferedImage
*** JpegByteImageSequence probably just needs Bitmap instead of BufferedImage
*** LoadFileImageSequance probably just needs Bitmap instead of BufferedImage
* testing
** BoofTesting needs Bitmap instead of BufferedImage

== Partial Android Port of BoofCV v0.2 ==
For his [[Projects_Using_BoofCV|master thesis]], [[User:Maarten_Van_Lier|Maarten]] made a partial port of BoofCV v0.2 to Android.
In this port, the following was done to get it working:
* Just use the EJML, GeoRegression and libpja libraries that BoofCV uses, they work on Android as is.
* Modify the core.image.ConvertBufferedImage class to the core.image.ConvertBitmap class. This new class uses BufferedImages and Android API calls to do the same as the ConvertBufferedImage class. The implementation is not as efficient as the original one, though.
* Leave out the gui package because of the java swing windows that were used. It was not needed for the thesis.
* Leave out problematic methods in io.image.UtilIO and remove other classes in io.image, io.video and io.wrapper.images, because they use java swing window calls. They were not needed for the thesis.
* Leave out the methods in testing.BoofTesting that used the BufferedImage class.

Basically, code for loading images was ported, while code to show windows was left out.
The code is available here (TODO: add link). Note that this is an incomplete port that may not have what you need, or may not work as expected.