scaffoldhunter-devel — Discussion list for Scaffold Hunter developers

Re: [Scaffoldhunter-devel] GSoC 2014 participation and Command Line Support project

[Check N. <che...@gm...>]

Hi,

I am currently writing my proposal, so I'll like to know the IRC
freenode channel where I could directly ask for my review and pressing
questions. I'll also love to know how I can create wiki pages on the
project website where I could post my proposal before applying
directly.

Cheers!
Nyah

Re: [Scaffoldhunter-devel] GSoC 2014 Molecule Clouds

[Till S. <til...@tu...>]

Hi, 
please allays ask your questions over the mailing list. That means, we do not need to repeat ourself for similar questions. 

Am Montag, 10. März 2014, 21:14:24 schrieb Anjenson:
> Hi again, Till.
> 
> Sorry for not replying for long time.
> I was looking for the way to build SH source code and I am just 
> wandering if anything like maven is used?
No. We have an ant build and a Eclipse project File.

Greetings 
Till 

> 
> Regards,
> Andrew ("Anjenson")
> On 03/04/2014 01:23 PM, Till Schäfer wrote:
> > Hi,
> > i propose the following way to familiarize yourself with the SH software:
> >
> > 1. Get an overview of the user side functionality.  You can use the tutorial [1] to do so. You do not need to understand the chemical details in depth at this point, but you should get what a view is, how the subsetbar works, how the selection works and the general concepts of the SH software.
> > 2. As you are interested in implementing a view, it would be natural to start with the View / GenericView  interface in the package edu.udo.scaffoldhunter.view. Figure out the purpose of the different methods.
> > 3. Get in touch with the database layer (edu.udo.scaffoldhunter.db). This layer contains the objects (molecules, scaffolds, subsets, etc) you will need to handle in your view. Most of them are POJO style classes. It is sufficient to understand how the different objects are connected/related to each other at this point. You will only get very seldom in contact with the DbManager class (which handles the loading / saving of the data).
> > 4. Familiarize yourself with Piccolo [2]. It is the toolkit we use for visualization. You can also use the other views as an example how to visualize the molecules/scaffolds in piccolo.
> >
> > I think this will take some time ;-) Feels free to ask further questions.
> >
> > [1] http://scaffoldhunter.sourceforge.net/tutorials.html
> > [2] http://www.piccolo2d.org/learn/get-started.html
> >
> > Greetings
> > Till
> >
> > Am Montag, 3. März 2014, 19:47:01 schrieb Anjenson:
> >> Hi,
> >> I am 3rd-year student of BSU FAMCS, department of Discrete Math and
> >> Algorithmics.
> >> I am interested in working for the "Molecule Clouds" project.
> >>
> >> I have a pretty good experience with Java (especially in Web
> >> Development) and some experience with creating GUI with java.swing and
> >> java.awt. However, I will try to do my best to learn as much about
> >> Piccolo2D as possible before the start of the program.
> >> I would appreciate any advice on how to start getting to know the codebase.
> >>
> >> Regards,
> >> Andrew Zhilka
> >>
> >> ------------------------------------------------------------------------------
> >> Subversion Kills Productivity. Get off Subversion & Make the Move to Perforce.
> >> With Perforce, you get hassle-free workflows. Merge that actually works.
> >> Faster operations. Version large binaries.  Built-in WAN optimization and the
> >> freedom to use Git, Perforce or both. Make the move to Perforce.
> >> http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk
> >> _______________________________________________
> >> Scaffoldhunter-devel mailing list
> >> Sca...@li...
> >> https://lists.sourceforge.net/lists/listinfo/scaffoldhunter-devel
> 
-- 
Dipl.-Inf. Till Schäfer
TU Dortmund University
Chair 11 - Algorithm Engineering
Otto-Hahn-Str. 14 / Room 237
44227 Dortmund, Germany

e-mail: til...@cs...
phone: +49(231)755-7706
fax: +49(231)755-7740
web: http://ls11-www.cs.uni-dortmund.de/staff/schaefer
pgp: https://keyserver2.pgp.com/vkd/SubmitSearch.event?&&SearchCriteria=0xD84DED79

Re: [Scaffoldhunter-devel] [GSoC 2014] Intoduction

[Tim W. <ti...@ho...>]

As stated previously I have little experience in a big context (only one large student project in Java). Nonetheless I've done private stuff in several languages also object orientated ones like Python 2.6, Dart and Ruby (on Rails). Python for coursework and Udacity tasks. C and C++ mostly in terms of SpoJ. Before coming across GSoC I was hooked with Go.Still, it doesn't mean I worked with them.
About the SAHNs, that's exactly what I meant. So the only difference is the independence of columns and rows, where the columns get clustered first and the rows afterwards.

I'll come back for coding questions after further investigation ;)
Regards,Tim

> From: til...@tu...
> To: sca...@li...
> CC: ti...@ho...
> Subject: Re: [Scaffoldhunter-devel] [GSoC 2014] Intoduction
> Date: Mon, 10 Mar 2014 11:20:55 +0100
> 
> Hi, 
> 
> Am Sonntag, 9. März 2014, 16:40:54 schrieb i Myself:
> > Hello,
> > My name is Tim Weidner. I'm a student of Computer Science at the University of Ulm (Germany), currently pursuing my bachelors degree. While looking for a suitable engagement in GSoC I stumbled upon ScaffoldHunter and was in love with the beautiful Java program. In the past I thought of Java as slow and ugly, but that's necessarily not the case as it seems. 
> Well, there is some overhead in some situations, but java is less error prone than c++ (especially for semi-professional programmers). Furthermore, algorithms might matter al lot more when dealing with computationally intensive tasks.
> > So I want to get involved in this, subscribed to the mailing lists, went through the GenericView/TreeMap packages, the tutorial and also the provided papers.Although it is still overwhelming to see the massive wall of code as an unexperienced (real world) developer, I think it's certainly not impossible for me with the right guidance. Sadly my work experience outside of university (Java) projects is non-existent, but that's why I'm here.As you might have guessed I'm particularly interested in the implementation of heatmaps. Coming from a bioinformatics background, I cannot say much about chemistry but certainly know what a dendrogram is or how clustering algorithms work. After reading the MPX Paper, I come to the conclusion that heatmaps in this context mainly differ by the fact that they allow more than two-way clustering, is that correct?
> What do you mean with "more than two way" clustering? The Paper uses two SAHN (Sequential Agglomerative Hierarchical Non-Overlapping) Clusterings (one for the Molecules/Objects and one for the Properties). The Clustering of the Molecules is independent of the properties displayed.> 
> > Two other meaningful questions:
> > Is there any special task we have to complete in order to be accepted as GSoC applicants?
> There will be a small Java-Test, after your application. 
> 
> > Why doesn't ScaffoldHunter utilize Git/Mercurial?
> Because we were always able to do our tasks efficiently with SVN (at least up to this point) and GIT was not very common the time this project stated. We have a very linear development and not so many branches. I also love to use some features of git like local commits. You can use an git/svn bridge for this purpose. 
> 
> > Digging into Piccolo2D now,feel free to ask me anything :)
> You mentioned that you have no Java-experience outside university. Have you worked with c++ or some other object oriented languages?
> No further questions at this point. Let me point the sentence back to you ;) Feel free to ask questions about the code structure, etc. We like to see that you are working with the code. 
> 
> Regards, 
> Till	 	   		  
> -- 
> Dipl.-Inf. Till Schäfer
> TU Dortmund University
> Chair 11 - Algorithm Engineering
> Otto-Hahn-Str. 14 / Room 237
> 44227 Dortmund, Germany
> 
> e-mail: til...@cs...
> phone: +49(231)755-7706
> fax: +49(231)755-7740
> web: http://ls11-www.cs.uni-dortmund.de/staff/schaefer
> pgp: https://keyserver2.pgp.com/vkd/SubmitSearch.event?&&SearchCriteria=0xD84DED79
 		 	   		  

Re: [Scaffoldhunter-devel] [GSoC 2014] Intoduction

[Till S. <til...@tu...>]

Hi, 

Am Sonntag, 9. März 2014, 16:40:54 schrieb i Myself:
> Hello,
> My name is Tim Weidner. I'm a student of Computer Science at the University of Ulm (Germany), currently pursuing my bachelors degree. While looking for a suitable engagement in GSoC I stumbled upon ScaffoldHunter and was in love with the beautiful Java program. In the past I thought of Java as slow and ugly, but that's necessarily not the case as it seems. 
Well, there is some overhead in some situations, but java is less error prone than c++ (especially for semi-professional programmers). Furthermore, algorithms might matter al lot more when dealing with computationally intensive tasks.
> So I want to get involved in this, subscribed to the mailing lists, went through the GenericView/TreeMap packages, the tutorial and also the provided papers.Although it is still overwhelming to see the massive wall of code as an unexperienced (real world) developer, I think it's certainly not impossible for me with the right guidance. Sadly my work experience outside of university (Java) projects is non-existent, but that's why I'm here.As you might have guessed I'm particularly interested in the implementation of heatmaps. Coming from a bioinformatics background, I cannot say much about chemistry but certainly know what a dendrogram is or how clustering algorithms work. After reading the MPX Paper, I come to the conclusion that heatmaps in this context mainly differ by the fact that they allow more than two-way clustering, is that correct?
What do you mean with "more than two way" clustering? The Paper uses two SAHN (Sequential Agglomerative Hierarchical Non-Overlapping) Clusterings (one for the Molecules/Objects and one for the Properties). The Clustering of the Molecules is independent of the properties displayed. 

> Two other meaningful questions:
> Is there any special task we have to complete in order to be accepted as GSoC applicants?
There will be a small Java-Test, after your application. 

> Why doesn't ScaffoldHunter utilize Git/Mercurial?
Because we were always able to do our tasks efficiently with SVN (at least up to this point) and GIT was not very common the time this project stated. We have a very linear development and not so many branches. I also love to use some features of git like local commits. You can use an git/svn bridge for this purpose. 

> Digging into Piccolo2D now,feel free to ask me anything :)
You mentioned that you have no Java-experience outside university. Have you worked with c++ or some other object oriented languages?
No further questions at this point. Let me point the sentence back to you ;) Feel free to ask questions about the code structure, etc. We like to see that you are working with the code. 

Regards, 
Till	 	   		  
-- 
Dipl.-Inf. Till Schäfer
TU Dortmund University
Chair 11 - Algorithm Engineering
Otto-Hahn-Str. 14 / Room 237
44227 Dortmund, Germany

e-mail: til...@cs...
phone: +49(231)755-7706
fax: +49(231)755-7740
web: http://ls11-www.cs.uni-dortmund.de/staff/schaefer
pgp: https://keyserver2.pgp.com/vkd/SubmitSearch.event?&&SearchCriteria=0xD84DED79

[Scaffoldhunter-devel] [GSoC 2014] Intoduction

[i M. <ti...@ho...>]

Hello,
My name is Tim Weidner. I'm a student of Computer Science at the University of Ulm (Germany), currently pursuing my bachelors degree. While looking for a suitable engagement in GSoC I stumbled upon ScaffoldHunter and was in love with the beautiful Java program. In the past I thought of Java as slow and ugly, but that's necessarily not the case as it seems. So I want to get involved in this, subscribed to the mailing lists, went through the GenericView/TreeMap packages, the tutorial and also the provided papers.Although it is still overwhelming to see the massive wall of code as an unexperienced (real world) developer, I think it's certainly not impossible for me with the right guidance. Sadly my work experience outside of university (Java) projects is non-existent, but that's why I'm here.As you might have guessed I'm particularly interested in the implementation of heatmaps. Coming from a bioinformatics background, I cannot say much about chemistry but certainly know what a dendrogram is or how clustering algorithms work. After reading the MPX Paper, I come to the conclusion that heatmaps in this context mainly differ by the fact that they allow more than two-way clustering, is that correct?
Two other meaningful questions:
Is there any special task we have to complete in order to be accepted as GSoC applicants?
Why doesn't ScaffoldHunter utilize Git/Mercurial?
Digging into Piccolo2D now,feel free to ask me anything :)
Regards,Tim 		 	   		  

Re: [Scaffoldhunter-devel] GSoC 2014- Visualization of Clustering Results, Ramachandran Plots and Heatmaps

[<nl...@us...>]

Hello Werner,

thanks for your interest in Scaffold Hunter. For this project idea you may 
also want to go through the packages edu.udo.scaffoldhunter.model.clustering 
and edu.udo.scaffoldhunter.view.dendrogram to see how hierarchical clustering 
and the dendrogram view are implemented. The literature linked at the project 
ideas page provides an example how heatmaps and clustering can be combined for 
an application in cheminformatics.

Since you are interested in visualization, the project idea "Molecule Clouds" 
might also be interesting for you. We already have obtained code from the 
authors who developed the approach, but it generated static images only. So 
using parts of the code and developing and interactive view for the approach 
also is quite a nice visualization project. You may use Google and search for 
(interactive) word/tag cloud to get an idea how an interactive visualization 
could look like.


Regards,
  Nils
 

On Wednesday 05 March 2014 13:18:35 Werner Sturm wrote:
> Hello,
> 
> My name is Werner Sturm and I study Computer Science (master's degree)
> at Technical University of Graz in Austria. My fields of interest are
> information visualization, computer graphics and software technology. I
> am mostly experienced in object oriented programming languages like Java
> and C++. I also have experience in JavaScript + HTML, C# and bash
> scripting. I've already done many internships as a Java software
> developer for a company over several years. Thus, I have experienced
> several frameworks/libraries like Hibernate, Swing, Apache Axis, etc.
> Moreover, I worked on a web-based scientific visualization (2D and 3D)
> of energy consumption for automotive engineering [1].
> 
> Now, I would like to tackle a new project about visualization to expand
> my experience in this field. The project idea "Visualization of
> Clustering Results, Ramachandran Plots and Heatmaps" caught my interest
> and so I would appreciate further information about this idea. My next
> step is going through the mentioned package
> "edu.udo.scaffoldhunter.view.treemap" in a more detailed way.
> 
> If you need more information about myself, don't hesitate to ask.
> 
> Best regards,
> Werner Sturm
> 
> [1]
> http://www.thinkmind.org/index.php?view=article&articleid=content_2013_1_10_
> 60019
> 
> 
> ----------------------------------------------------------------------------
> -- Subversion Kills Productivity. Get off Subversion & Make the Move to
> Perforce. With Perforce, you get hassle-free workflows. Merge that actually
> works. Faster operations. Version large binaries.  Built-in WAN
> optimization and the freedom to use Git, Perforce or both. Make the move to
> Perforce.
> http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk
> _______________________________________________
> Scaffoldhunter-devel mailing list
> Sca...@li...
> https://lists.sourceforge.net/lists/listinfo/scaffoldhunter-devel

Re: [Scaffoldhunter-devel] GSoC 2014 participation and Command Line Support project

[Check N. <che...@gm...>]

Sorry for the late reply nils,

I was writing exams, now i'm done ^_^. would take a look at the
tutorials and get and get an svn checkout and start writing up my GSoC
proposal for some early review and see what I could start working on
at the moment.

Regards,
Nyah

On 3/3/14, nl...@us... <nl...@us...> wrote:
> Hi Nyah,
>
> thanks for your interest. A good starting point is to look at the tutorial
> [1]
> to make yourself familiar with the software and the analysis methods
> supported
> by Scaffold Hunter. You can check out the source code from the SVN
> repository
> (trunk). The packages edu.udo.scaffoldhunter.model.clustering and
> edu.udo.scaffoldhunter.model.treegen are most important for the project idea
>
> since they contain the code for clustering and the scaffold tree generation,
>
> which should be supported by a CLI. We have recently added a very basic
> command line tool for a specific task that can be found in the package
> edu.udo.scaffoldhunter.cli. It uses the library JCommander [2] for argument
>
> parsing. Of course, you are free to use a different library. However, you
> may
> want to have a look at the code to get an idea how command line support for
>
> Scaffold Hunter can be implemented.
>
>
> Regards,
>   Nils
>
> [1] http://scaffoldhunter.sourceforge.net/tutorials.html
> [2] http://jcommander.org/
>
>
> On Thursday 27 February 2014 13:46:16 Check Nyah wrote:
>> Hi,
>> I am Nyah Check, a sophomore computer engineering student at the
>> University of Buea(Cameroon), a 3 year Biochemistry student at my
>> University where I did some deep study of nucleotides and cheminformatics
>> which is this project's application domain and a GSoC participant for
>> 2013
>> where I worked with BRL-CAD on implementing a pull routine to reverse
>> the effects
>> of a push on Geometry. I wish to implement the command line interface for
>> this great software and would love to get some guidance on where to look
>> in
>> the software, online tutorials I could look at in addition to the
>> Scaffold
>> preprint document which would assist me in implementing this project and
>> alsothe priority of this project in relation to the organization's vision
>> for this year's GSoC. I see this project as both an intellectual
>> challenge
>> and a forum to exercise my Java and also interact with the cutting edge
>> researchers
>> in this domain. Can't wait to work with you guys this year :-). Thanks
>> for
>> the assistance.
>>
>> Cheers!
>> Nyah
>

[Scaffoldhunter-devel] GSoC 2014- Visualization of Clustering Results, Ramachandran Plots and Heatmaps

[Werner S. <wer...@gm...>]

Hello,

My name is Werner Sturm and I study Computer Science (master's degree) 
at Technical University of Graz in Austria. My fields of interest are 
information visualization, computer graphics and software technology. I 
am mostly experienced in object oriented programming languages like Java 
and C++. I also have experience in JavaScript + HTML, C# and bash 
scripting. I've already done many internships as a Java software 
developer for a company over several years. Thus, I have experienced 
several frameworks/libraries like Hibernate, Swing, Apache Axis, etc.  
Moreover, I worked on a web-based scientific visualization (2D and 3D) 
of energy consumption for automotive engineering [1].

Now, I would like to tackle a new project about visualization to expand 
my experience in this field. The project idea "Visualization of 
Clustering Results, Ramachandran Plots and Heatmaps" caught my interest 
and so I would appreciate further information about this idea. My next 
step is going through the mentioned package 
"edu.udo.scaffoldhunter.view.treemap" in a more detailed way.

If you need more information about myself, don't hesitate to ask.

Best regards,
Werner Sturm

[1] 
http://www.thinkmind.org/index.php?view=article&articleid=content_2013_1_10_60019 

Re: [Scaffoldhunter-devel] GSoC 2014 Molecule Clouds

[Till S. <til...@tu...>]

Hi,
i propose the following way to familiarize yourself with the SH software: 

1. Get an overview of the user side functionality.  You can use the tutorial [1] to do so. You do not need to understand the chemical details in depth at this point, but you should get what a view is, how the subsetbar works, how the selection works and the general concepts of the SH software. 
2. As you are interested in implementing a view, it would be natural to start with the View / GenericView  interface in the package edu.udo.scaffoldhunter.view. Figure out the purpose of the different methods. 
3. Get in touch with the database layer (edu.udo.scaffoldhunter.db). This layer contains the objects (molecules, scaffolds, subsets, etc) you will need to handle in your view. Most of them are POJO style classes. It is sufficient to understand how the different objects are connected/related to each other at this point. You will only get very seldom in contact with the DbManager class (which handles the loading / saving of the data). 
4. Familiarize yourself with Piccolo [2]. It is the toolkit we use for visualization. You can also use the other views as an example how to visualize the molecules/scaffolds in piccolo.

I think this will take some time ;-) Feels free to ask further questions. 

[1] http://scaffoldhunter.sourceforge.net/tutorials.html
[2] http://www.piccolo2d.org/learn/get-started.html

Greetings 
Till

Am Montag, 3. März 2014, 19:47:01 schrieb Anjenson:
> Hi,
> I am 3rd-year student of BSU FAMCS, department of Discrete Math and 
> Algorithmics.
> I am interested in working for the "Molecule Clouds" project.
> 
> I have a pretty good experience with Java (especially in Web 
> Development) and some experience with creating GUI with java.swing and 
> java.awt. However, I will try to do my best to learn as much about 
> Piccolo2D as possible before the start of the program.
> I would appreciate any advice on how to start getting to know the codebase.
> 
> Regards,
> Andrew Zhilka
> 
> ------------------------------------------------------------------------------
> Subversion Kills Productivity. Get off Subversion & Make the Move to Perforce.
> With Perforce, you get hassle-free workflows. Merge that actually works. 
> Faster operations. Version large binaries.  Built-in WAN optimization and the
> freedom to use Git, Perforce or both. Make the move to Perforce.
> http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk
> _______________________________________________
> Scaffoldhunter-devel mailing list
> Sca...@li...
> https://lists.sourceforge.net/lists/listinfo/scaffoldhunter-devel
-- 
Dipl.-Inf. Till Schäfer
TU Dortmund University
Chair 11 - Algorithm Engineering
Otto-Hahn-Str. 14 / Room 237
44227 Dortmund, Germany

e-mail: til...@cs...
phone: +49(231)755-7706
fax: +49(231)755-7740
web: http://ls11-www.cs.uni-dortmund.de/staff/schaefer
pgp: https://keyserver2.pgp.com/vkd/SubmitSearch.event?&&SearchCriteria=0xD84DED79

[Scaffoldhunter-devel] GSoC 2014 Molecule Clouds

[Anjenson <hot...@gm...>]

Hi,
I am 3rd-year student of BSU FAMCS, department of Discrete Math and 
Algorithmics.
I am interested in working for the "Molecule Clouds" project.

I have a pretty good experience with Java (especially in Web 
Development) and some experience with creating GUI with java.swing and 
java.awt. However, I will try to do my best to learn as much about 
Piccolo2D as possible before the start of the program.
I would appreciate any advice on how to start getting to know the codebase.

Regards,
Andrew Zhilka

Re: [Scaffoldhunter-devel] GSoC 2014-"Visualization of Clustering Results, Ramachandran Plots and Heatmaps"

[<nl...@us...>]

Hi Abhinav Mittal,

good knowledge of Java (not necessarily Netbeans) is an important requirement 
for the project. I think it is very difficult to make oneself familiar with 
the existing code base and, especially, to implement a new view, when having 
only little experience with Java. However, since you have good experience with 
other programming languages, you might want to have a look at the code from 
the repository and give it a try. The package 
edu.udo.scaffoldhunter.view.treemap should be interesting for the project.


Regards,
  Nils


On Sunday 02 March 2014 08:39:43 Abhinav Mittal wrote:
> Hi
> My name is Abhinav Mittal. I am 3rd year student at IIIT Hyderabad pursuing
> Computational Natural Sciences. I am interested in working for Scaffold
> Hunter on above mentioned idea for GSoC 2014.
> I have good experience with javascript, python, c , MySQL and scientific
> software packages like PyMol,VMD-NAMD and GAMESS(US) but little with
> JAVA-NETBEANS.  Is Scaffold Hunter a good option for me from GSoC 2014
> perspective?
> I am really interested in this kind of stuff and already going through the
> mentioned research papers.
> I am presently working on structural bioinformatics of RNA under Prof.
> Abhijit Mitra(http://faculty.iiit.ac.in/~abi_chem/) at Center of
> Computational Natural Sciences and Bioinformatics, IIIT Hyderbad.

Re: [Scaffoldhunter-devel] GSoC 2014 participation and Command Line Support project

[<nl...@us...>]

Hi Nyah,

thanks for your interest. A good starting point is to look at the tutorial [1] 
to make yourself familiar with the software and the analysis methods supported 
by Scaffold Hunter. You can check out the source code from the SVN repository 
(trunk). The packages edu.udo.scaffoldhunter.model.clustering and 
edu.udo.scaffoldhunter.model.treegen are most important for the project idea 
since they contain the code for clustering and the scaffold tree generation, 
which should be supported by a CLI. We have recently added a very basic 
command line tool for a specific task that can be found in the package 
edu.udo.scaffoldhunter.cli. It uses the library JCommander [2] for argument 
parsing. Of course, you are free to use a different library. However, you may 
want to have a look at the code to get an idea how command line support for 
Scaffold Hunter can be implemented.


Regards,
  Nils

[1] http://scaffoldhunter.sourceforge.net/tutorials.html
[2] http://jcommander.org/


On Thursday 27 February 2014 13:46:16 Check Nyah wrote:
> Hi,
> I am Nyah Check, a sophomore computer engineering student at the
> University of Buea(Cameroon), a 3 year Biochemistry student at my
> University where I did some deep study of nucleotides and cheminformatics
> which is this project's application domain and a GSoC participant for 2013
> where I worked with BRL-CAD on implementing a pull routine to reverse
> the effects
> of a push on Geometry. I wish to implement the command line interface for
> this great software and would love to get some guidance on where to look in
> the software, online tutorials I could look at in addition to the Scaffold
> preprint document which would assist me in implementing this project and
> alsothe priority of this project in relation to the organization's vision
> for this year's GSoC. I see this project as both an intellectual challenge
> and a forum to exercise my Java and also interact with the cutting edge
> researchers
> in this domain. Can't wait to work with you guys this year :-). Thanks for
> the assistance.
> 
> Cheers!
> Nyah

[Scaffoldhunter-devel] GSoC 2014-"Visualization of Clustering Results, Ramachandran Plots and Heatmaps"

[Abhinav M. <apo...@gm...>]

Hi
My name is Abhinav Mittal. I am 3rd year student at IIIT Hyderabad pursuing
Computational Natural Sciences. I am interested in working for Scaffold
Hunter on above mentioned idea for GSoC 2014.
I have good experience with javascript, python, c , MySQL and scientific
software packages like PyMol,VMD-NAMD and GAMESS(US) but little with
JAVA-NETBEANS.  Is Scaffold Hunter a good option for me from GSoC 2014
perspective?
I am really interested in this kind of stuff and already going through the
mentioned research papers.
I am presently working on structural bioinformatics of RNA under Prof.
Abhijit Mitra(http://faculty.iiit.ac.in/~abi_chem/) at Center of
Computational Natural Sciences and Bioinformatics, IIIT Hyderbad.

[Scaffoldhunter-devel] GSoC 2014 participation and Command Line Support project

[Check N. <che...@gm...>]

Hi,
I am Nyah Check, a sophomore computer engineering student at the
University of Buea(Cameroon), a 3 year Biochemistry student at my
University where I did some deep study of nucleotides and cheminformatics
which is this project's application domain and a GSoC participant for 2013
where I worked with BRL-CAD on implementing a pull routine to reverse
the effects
of a push on Geometry. I wish to implement the command line interface for
this great software and would love to get some guidance on where to look in
the software, online tutorials I could look at in addition to the Scaffold
preprint document which would assist me in implementing this project and
alsothe priority of this project in relation to the organization's vision
for this year's GSoC. I see this project as both an intellectual challenge
and a forum to exercise my Java and also interact with the cutting edge
researchers
in this domain. Can't wait to work with you guys this year :-). Thanks for
the assistance.

Cheers!
Nyah

[Scaffoldhunter-devel] Making releases more transparent

[Till S. <til...@tu...>]

Hi, 
i would like to propose a more transparent way for release planing, as the main contributers tend to be more distributed all over the world. I suppose to use the bug tracker more extensively for release planing:

I already introduced the new milestones in a previous mail (see attachment). We now tag a bug with "any future version" by default and move this bug to a specific version if we fix the bug or do a release planing. However, it has the drawback, that we do not always know if the next release may be 2.3.1 (only bug fixing) or 2.4.0 (feature release) if we fix a bug. Therefore, we introduce a new Milestone "next release", which should be selected in this case.

If we want to release a new version, there should be a bug report with the title "Scaffold Hunter version X.Y.Z release request". This should give all participants the possibility to discuss show blockers, that should be integrated into the next release and also to determine if we do a feature or bugfix release. We can then simply rename the milestone "next release" to the concrete new version. 

What do you think about this?

Greetings
Till

-- 
Dipl.-Inf. Till Schäfer
TU Dortmund University
Chair 11 - Algorithm Engineering
Otto-Hahn-Str. 14 / Room 237
44227 Dortmund, Germany

e-mail: til...@cs...
phone: +49(231)755-7706
fax: +49(231)755-7740
web: http://ls11-www.cs.uni-dortmund.de/staff/schaefer
pgp: https://keyserver2.pgp.com/vkd/SubmitSearch.event?&&SearchCriteria=0xD84DED79

[Scaffoldhunter-devel] Bug/Wish-Tracker - New Milestones

[Till S. <til...@tu...>]

Hi everyone,
in the source forge tracker for bugs and wishes, we had a mess of different milestones. Many bugs were assigned to milestone which were released long time ago. We used to have milestones like "2.2.x", which leads to the situation, that old bugs that are not fixed in 2.2.x have to be manually updated to 2.3.x. As SF does not support batch processing, this was nearly never done. 

I have created a new milestone "any future version" in the sf-tracker, which should be used for any new bug and wish. This will avoid the above situation, where the bugs have to be moved to new milestones. 

We will move the bugs to a specific version now only for concrete planing. That means, if we want to release a new version and are planing which bugs should be fixed before the release, we will move this bug to the specific version. 

Sorry to the peoples that are receiving change notification for the bugs and have received all that change spam from reassigning every single bug to a new milestone.

Greetings 
Till 

-- 
Dipl.-Inf. Till Schäfer
Technische Universität Dortmund
Chair 11 - Algorithm Engineering
Otto-Hahn-Str. 14 / Raum 237
44227 Dortmund, Germany

e-mail: til...@cs...
phone: +49(231)755-7706
fax: +49(231)755-7740
web: http://ls11-www.cs.uni-dortmund.de/staff/schaefer
pgp: https://keyserver2.pgp.com/vkd/SubmitSearch.event?&&SearchCriteria=0xD84DED79

[Scaffoldhunter-devel] Creating button icons

[Falk N. <ne...@mi...>]

Hi everyone,

what software was used to create the icons SH uses for the toolbar buttons?

Thanks,
Falk

[Scaffoldhunter-devel] GSoC accepted students

[Karsten K. <ka...@us...>]

Hi,

Google has announced the accepted students for this year's summer of
code. We have two accepted students,

Falk Nette (Visual Feature selection)
Jeroen Lappenschaar (Ramachandran Plots, Treemaps, and Heatmaps)

Congratulations to Falk and Jeroen, we will contact you soon
to discuss the further process.

For all the others who tried, thanks for your interest in the Scaffold
Hunter projects. We had a couple of further good proposals, but only
two slots to assign. You are welcome to join the project and participate
in the development of Scaffold Hunter outside of GSoC.

Best,
    Karsten

Re: [Scaffoldhunter-devel] Scaffold Hunter Java Test

[Till S. <til...@tu...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,
the code looks good, too ;-)

Till

Am 27.05.2013 08:52, schrieb Karsten Klein:
> Hi Jeroen,
> 
> thanks for sending that through. A first empirical evaluation shows that indeed the plugin is loaded and at least
> some calculation is performed ;-)
> 
> Best, Karsten
> 
> On 25/05/2013 8:07 AM, Jeroen Lappenschaar wrote:
>> Hi,
>> 
>> I hope you have found the previous code in good condition.
>> 
>> In the attachment is the SH plugin you requested as a test. I know it was not a requirement anymore, but I had
>> the evening off and I liked to show my good intentions.
>> 
>> (The JAR is just the given java files in the right package)
>> 
>> Jeroen
>> 
>> 
>> 2013/5/16 Jeroen Lappenschaar <lap...@gm... <mailto:lap...@gm...>>
>> 
>> Hi there fellow developers,
>> 
>> I'm giving you hereby my answers to the Java Test that I got from Till Schäfer. My answers to the SH test will
>> follow since I'm typing this from my vacation adress in not-yet-so-sunny Barcelona on a Spanish keyboard.
>> 
>> //Due to my vacation adress the following code is not compiled. It's unfortunate but it might even give you some
>> more insight ;)
>> 
>> With kind regards, Jeroen Lappenschaar applying for the Google Summer of Code
>> 
>> 
>> ************ *Java Tasks* ************
>> 
>> *** Off Mean *** Complete the following method, which calculates the mean for all odd indices of the array.
>> 
>> /** * Returns the mean for all odd indices of the given array. * Returns 0.0 if array size is 0. */ public float
>> getOddMean(float[] numbers) { if(numbers.length == 0) return 0.0; float mean = 0.0; int j = 0; for(int i = 1; i <
>> numbers.length; i+=2){ mean += numbers[i]; j++; } return mean/j; }
>> 
>> 
>> *** End Contracts *** Complete the following method of the class 'Division'. It should remove all employees,
>> which contract ends before the the given date and return them as list.
>> 
>> /** * Removes all employees which contract ends before the given date and returns these employees */ public
>> List<Employee> endContracts(Date date) { List<Employee> removed = new ArrayList<Employee>(); for(Employee
>> employee : employees) if(employee.getEndOfContract().before(date)) { removed.add(employee); 
>> employees.remove(employee); //this uses the hash method described below } return removed; }
>> 
>> 
>> *** Protect Employees *** The method 'getEmployees' allows to manipulate the content of the private attribute
>> 'employees'. Modify the method 'getEmployees' in such a way that this is not possible anymore.
>> 
>> /** * Returns an *unmoditiable* list of employees */ public List<Employee> getEmployees() { return
>> Collections.unmodifiableList(employees); }
>> 
>> *** Change monthlyWage *** Add a method to modify the monthly wage of an employee. Make sure that the class
>> 'Division' is the only class with access to this method. Note that 'Division' is the only other class in the 
>> package of the class 'Employee'.
>> 
>> //package private void setMonthlyWage(int newWage) { montlyWage = newWage; }
>> 
>> 
>> *** Equal Employees *** Provide a standard way to check the equality of two employees. Two employees should be
>> equal iff the personnelNumber is equal.
>> 
>> public boolean equals(Object object) { if(! object instanceof Employee) return false; Employee employee =
>> (Employee) object; return employee.getPersonnelNumber() == personnelNumber(); }
>> 
>> //together with equals we also change the hashcode function public int hashCode() { return personnelNumber; }
>> 
>> *** Generics *** Change the class node in such a way that it can store a content of arbitrary type. Make sure
>> that the all nodes in a connected tree can only have the same content type.
>> 
>> public class Node<E> {
>> 
>> private Node<E> parent; private List<Node<E>> children; private E content;
>> 
>> public Node(E content) { this.children = new ArrayList<Node<E>>(); this.content = content; }
>> 
>> public void addChild(Node<E> newChild) { newChild.parent = this; children.add(newChild); }
>> 
>> public E getContent() { return content; } }
>> 
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------ Try New Relic Now & We'll Send You
>> this Cool Shirt New Relic is the only SaaS-based application performance monitoring service that delivers
>> powerful full stack analytics. Optimize and monitor your browser, app, & servers with just a few lines of code.
>> Try New Relic and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
>> 
>> 
>> _______________________________________________ Scaffoldhunter-devel mailing list 
>> Sca...@li... https://lists.sourceforge.net/lists/listinfo/scaffoldhunter-devel
> 
> 
> ------------------------------------------------------------------------------ Try New Relic Now & We'll Send You
> this Cool Shirt New Relic is the only SaaS-based application performance monitoring service that delivers powerful
> full stack analytics. Optimize and monitor your browser, app, & servers with just a few lines of code. Try New
> Relic and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may 
> _______________________________________________ Scaffoldhunter-devel mailing list 
> Sca...@li... https://lists.sourceforge.net/lists/listinfo/scaffoldhunter-devel
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=ldIh
-----END PGP SIGNATURE-----

Re: [Scaffoldhunter-devel] Scaffold Hunter Java Test

[Karsten K. <kar...@tu...>]

Hi Jeroen,

thanks for sending that through. A first empirical evaluation shows that
indeed the plugin is loaded and at least some calculation is performed ;-)

Best,
    Karsten

On 25/05/2013 8:07 AM, Jeroen Lappenschaar wrote:
> Hi, 
>
> I hope you have found the previous code in good condition.
>
> In the attachment is the SH plugin you requested as a test. I know it
> was not a requirement anymore, but I had the evening off and I liked
> to show my good intentions. 
>
> (The JAR is just the given java files in the right package)
>
> Jeroen
>
>
> 2013/5/16 Jeroen Lappenschaar <lap...@gm...
> <mailto:lap...@gm...>>
>
>     Hi there fellow developers, 
>
>     I'm giving you hereby my answers to the Java Test that I got from
>     Till Schäfer. My answers to the SH test will follow since I'm
>     typing this from my vacation adress in not-yet-so-sunny Barcelona
>     on a Spanish keyboard. 
>
>     //Due to my vacation adress the following code is not compiled.
>     It's unfortunate but it might even give you some more insight ;)
>
>     With kind regards, 
>     Jeroen Lappenschaar
>     applying for the Google Summer of Code
>
>
>     ************
>     *Java Tasks*
>     ************
>
>     *** Off Mean ***
>     Complete the following method, which calculates the mean for all
>     odd indices of the array.
>
>     /**
>      * Returns the mean for all odd indices of the given array. 
>      * Returns 0.0 if array size is 0. 
>      */
>     public float getOddMean(float[] numbers) {
>     if(numbers.length == 0)
>     return 0.0;
>     float mean = 0.0; 
>     int j = 0;
>     for(int i = 1; i < numbers.length; i+=2){
>     mean += numbers[i];
>     j++;
>     }
>     return mean/j;
>     }
>
>
>     *** End Contracts ***
>     Complete the following method of the class 'Division'. It should
>     remove all employees, which contract ends before the the given
>     date and return them as list.
>
>     /**
>      * Removes all employees which contract ends before the given date
>     and returns these employees
>      */
>     public List<Employee> endContracts(Date date) {
>     List<Employee> removed = new ArrayList<Employee>();
>     for(Employee employee : employees)
>     if(employee.getEndOfContract().before(date)) {
>     removed.add(employee);
>     employees.remove(employee); //this uses the hash method described
>     below
>     }
>     return removed;
>     }
>
>
>     *** Protect Employees ***
>     The method 'getEmployees' allows to manipulate the content of the
>     private attribute 'employees'. Modify the method 'getEmployees' in
>     such a way that this is not possible anymore.
>
>     /**
>      * Returns an *unmoditiable* list of employees
>      */
>     public List<Employee> getEmployees() {
>     return Collections.unmodifiableList(employees);
>     }
>
>     *** Change monthlyWage ***
>     Add a method to modify the monthly wage of an employee. Make sure
>     that the class 'Division' is the only class with access to this
>     method. Note that 'Division' is the only other class in the
>     package of the class 'Employee'.
>
>     //package private
>     void setMonthlyWage(int newWage) {
>     montlyWage = newWage;
>     }
>
>
>     *** Equal Employees ***
>     Provide a standard way to check the equality of two employees. Two
>     employees should be equal iff the personnelNumber is equal.
>
>     public boolean equals(Object object) {
>     if(! object instanceof Employee)
>     return false;
>     Employee employee = (Employee) object;
>     return employee.getPersonnelNumber() == personnelNumber();
>     }
>
>     //together with equals we also change the hashcode function
>     public int hashCode() {
>     return personnelNumber;
>     }
>
>     *** Generics ***
>     Change the class node in such a way that it can store a content of
>     arbitrary type. Make sure that the all nodes in a connected tree
>     can only have the same content type.
>
>     public class Node<E> {
>
>       private Node<E> parent;
>       private List<Node<E>> children;
>       private E content;
>
>       public Node(E content) {
>         this.children = new ArrayList<Node<E>>();
>         this.content = content;
>       }
>
>       public void addChild(Node<E> newChild) {
>         newChild.parent = this;
>         children.add(newChild);
>       }
>
>       public E getContent() {
>         return content;
>       }
>     }
>
>
>
>
>
> ------------------------------------------------------------------------------
> Try New Relic Now & We'll Send You this Cool Shirt
> New Relic is the only SaaS-based application performance monitoring service 
> that delivers powerful full stack analytics. Optimize and monitor your
> browser, app, & servers with just a few lines of code. Try New Relic
> and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
>
>
> _______________________________________________
> Scaffoldhunter-devel mailing list
> Sca...@li...
> https://lists.sourceforge.net/lists/listinfo/scaffoldhunter-devel

[Scaffoldhunter-devel] Scaffold Hunter Java Test

[Jeroen L. <lap...@gm...>]

Hi there fellow developers,

I'm giving you hereby my answers to the Java Test that I got from Till
Schäfer. My answers to the SH test will follow since I'm typing this from
my vacation adress in not-yet-so-sunny Barcelona on a Spanish keyboard.

//Due to my vacation adress the following code is not compiled. It's
unfortunate but it might even give you some more insight ;)

With kind regards,
Jeroen Lappenschaar
applying for the Google Summer of Code


************
*Java Tasks*
************

*** Off Mean ***
Complete the following method, which calculates the mean for all odd
indices of the array.

/**
 * Returns the mean for all odd indices of the given array.
 * Returns 0.0 if array size is 0.
 */
public float getOddMean(float[] numbers) {
if(numbers.length == 0)
return 0.0;
 float mean = 0.0;
int j = 0;
for(int i = 1; i < numbers.length; i+=2){
 mean += numbers[i];
j++;
}
return mean/j;
}


*** End Contracts ***
Complete the following method of the class 'Division'. It should remove all
employees, which contract ends before the the given date and return them as
list.

/**
 * Removes all employees which contract ends before the given date and
returns these employees
 */
public List<Employee> endContracts(Date date) {
List<Employee> removed = new ArrayList<Employee>();
 for(Employee employee : employees)
if(employee.getEndOfContract().before(date)) {
removed.add(employee);
 employees.remove(employee); //this uses the hash method described below
}
return removed;
}


*** Protect Employees ***
The method 'getEmployees' allows to manipulate the content of the private
attribute 'employees'. Modify the method 'getEmployees' in such a way that
this is not possible anymore.

/**
 * Returns an *unmoditiable* list of employees
 */
public List<Employee> getEmployees() {
return Collections.unmodifiableList(employees);
}

*** Change monthlyWage ***
Add a method to modify the monthly wage of an employee. Make sure that the
class 'Division' is the only class with access to this method. Note that
'Division' is the only other class in the package of the class 'Employee'.

//package private
void setMonthlyWage(int newWage) {
montlyWage = newWage;
}


*** Equal Employees ***
Provide a standard way to check the equality of two employees. Two
employees should be equal iff the personnelNumber is equal.

public boolean equals(Object object) {
if(! object instanceof Employee)
 return false;
Employee employee = (Employee) object;
return employee.getPersonnelNumber() == personnelNumber();
}

//together with equals we also change the hashcode function
public int hashCode() {
return personnelNumber;
}

*** Generics ***
Change the class node in such a way that it can store a content of
arbitrary type. Make sure that the all nodes in a connected tree can only
have the same content type.

public class Node<E> {

  private Node<E> parent;
  private List<Node<E>> children;
  private E content;

  public Node(E content) {
    this.children = new ArrayList<Node<E>>();
    this.content = content;
  }

  public void addChild(Node<E> newChild) {
    newChild.parent = this;
    children.add(newChild);
  }

  public E getContent() {
    return content;
  }
}

[Scaffoldhunter-devel] Google Summer of Code

[Karsten K. <kar...@ud...>]

Thanks to all students and mentors who participate in GSoC and want to
contribute to our project.

We received 19 proposals and will now go through them and rank them to
see which ones are
the most promising for us. Then we will have to see how many slots we
get from Google. Let's
hope for the best!
Acceptance will be announced on May 27 by Google.

Best,
    Karsten (on behalf of the Scaffold Hunter team)

Re: [Scaffoldhunter-devel] GSoC'13 proposal discussion

[Dushyant G. <goy...@gm...>]

Hi,

Thanks a lot for briefing me more about the project.
I have submitted my proposal.
Thanks.
Dushyant


On Tue, Apr 30, 2013 at 7:42 AM, Ashnil Kumar <as...@it...> wrote:

> Hi Dushyant,
>
> The intent of these projects is to develop a method for creating queries
> and for analysing retrieved images using different feature priorities. In
> 'Interactive Image Query Formulation' we could create a tool, as an
> extension to Scaffold Hunter, that allows the user to apply different image
> processing algorithms to construct his query, e.g. segmentation to select
> different regions of interest, interactively building a graph-based query.
> In 'Image Visual Feature Selection', we wish to extend Scaffold Hunter to
> analyse the retrieved images based on different feature sets/graph
> structures that are interactively adjusted to best fit the user's intent.
>
> The idea behind both these projects is that we do not assume that there is
> a "best" feature set; instead, the query and features are user dependent.
> As such, an interaction method is necessary for analysis and feedback to
> the retrieval algorithm.
>
> **
>
> We intend to use the LIDC data set (link below). You will be dealing with
> mainly 3D CT images. The feature set will be varied depending on the user
> but you can expect to deal with standard image features for describing
> texture, region of interest properties etc. We are also interested in
> extracting spatial relationships between regions of interest and
> representing these as graphs.
>
> Hope this helps.
>
> Regards,
> Ashnil
>
>
> ------------------------------------------------------------------------------
> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
> Get 100% visibility into your production application - at no cost.
> Code-level diagnostics for performance bottlenecks with <2% overhead
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap1
> _______________________________________________
> Scaffoldhunter-devel mailing list
> Sca...@li...
> https://lists.sourceforge.net/lists/listinfo/scaffoldhunter-devel
>
>

Re: [Scaffoldhunter-devel] GSoC: Scaffold Hunter for medical image retrieval

[<ne...@in...>]

Hi,

I have now uploaded my proposal to
https://google-melange.appspot.com/gsoc/proposal/review/google/gsoc2013/fnette/1

Comments/suggestions/questions welcomed.. Thank you for the advice given
so far!

Best regards,
Falk Nette

Re: [Scaffoldhunter-devel] GSOC project: Interactive Image Query Formation

[Jinman K. <jin...@sy...>]

Hi Girish,

I am starting to better understand your idea - I think I misunderstood you in my initial response.... In any case, my point about the difference between retrieval/classification is that if we want to find the most relevant image from the database to the query, we cant assume that there is a class (or a training sample) that the query image will relate to, at least not optimally. Exact or Approximate matching, as you describe it, however would work

As for your feature selection, your approach to defining classifiers is interesting and would provide real-time calculations. How flexible it is and how to ensure proper classes are defined needs to be thought about. In this project, we will use medical images and hence you should look into how features are used in e.g., CT data, see for example L. Dettori and L. Semler, 'A comparison of wavelet, ridgelet, and curvelet-based texture classification algorithms in computed tomogarphy' Computers in Biology and Medicine, 2007

Could you, in your proposal, elaborate on how your approach would fit into the Scaffold Hunter framework?

Best wishes.
Jinman

From: Girish Malkarnenkar [mailto:gir...@gm...]
Sent: Friday, 26 April 2013 3:52 AM
To: Jinman Kim
Cc: sca...@li...
Subject: Re: [Scaffoldhunter-devel] GSOC project: Interactive Image Query Formation

Hi Jinman,
Thanks for the feedback! :)

Regarding retrieval v/s classification, I meant we could perform retrieval the same way as done in nearest neighbor (either exact or approximate) based classification where we first find the most similar image. We can skip the classification step (where we would label the query image with the class of the most similar image) as we are interested only in finding the most similar image(s). I missed your point, about it being impossible to translate classification to retrieval. Can you explain a bit more?
Regarding the real time feature selection, I was thinking we could perhaps train different classifiers for the various features and at run time, depending on the user's input, we run only the necessary classifiers. Since each of the classifiers would be based on a super fast binary hashing (hamming distance) based calculation, we can guarantee real time performance for any of the classifiers or a combination of them (if the user selects similarity based on 2 features say). What do you feel?

Thanks,
Girish


On Thu, Apr 25, 2013 at 11:15 AM, Jinman Kim <jin...@sy...<mailto:jin...@sy...>> wrote:
Hi Girish,

Thanks for your interest and sharing your funny slides - gold membership is some achievement! Its good to see you applying state-of-the-art algorithms. I think you can build up to a strong proposal. Your proposal is interesting - it is impossible to translate classification to retrieval - may I suggest you explore these two points?

* In this project, we want to in real-time change the feature to use in the retrieval e.g. a dynamic feature selection / wieght
* Your idea of coarse-to-fine filtering could work - one could use classifier to narrow the large database and then use more conventional image-to-image matching algorithms....

best

Jinman
________________________________
From: Girish Malkarnenkar [gir...@gm...<mailto:gir...@gm...>]
Sent: Wednesday, 24 April 2013 2:06 AM
To: sca...@li...<mailto:sca...@li...>
Subject: [Scaffoldhunter-devel] GSOC project: Interactive Image Query Formation
Hi,

I found the description of this project<http://scaffoldhunter.sourceforge.net/wiki/doku.php?id=project_ideas#image_visual_feature_selection> pretty interesting. During my summer internship last year, I implemented a supervised binary hashing based approach for a database retrieval system for pose estimation using simple nearest neighbors. This project involved implementing state-of-the-art ideas from 3 IEEE CVPR 2012 (Computer Vision and Pattern Recognition) papers which ranged from the state of the art methods for locality sensitive hashing to fast nearest neighbor search by non-linear embedding and fast search in hamming space with multi index hashing. A short presentation about my implementation can be seen here<https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsdGRvbWFpbnxnaXJpc2htYWxrYXJuZW5rYXJ8Z3g6N2I4YjZkYTQ1YjA0NTFkMQ>.
I was wondering if a similar approach could be used here where the problem is formulated as a image retrieval task rather than a classification task. Since we want a real time querying experience, a binary hashing method (such as Spherical Hashing<http://sglab.kaist.ac.kr/Spherical_Hashing/>) might be very useful. Assuming the task to be finding the most similar image from a database of N images, we can train the binary hashing functions on the features (we can use either GIST<http://people.csail.mit.edu/torralba/code/spatialenvelope/>, SIFT<http://www.cs.ubc.ca/~lowe/keypoints/>, SURF<http://www.vision.ee.ethz.ch/~surf/> or a high dimensional combination of these using a bag of visual words approach). The binary hashing training algorithm will take care of figuring out which are the important dimensions/features from this combination of features. Given the high dimensional representation of each image, apart from using binary hashing which is a form of approximate k-Nearest Neighbors, we can also use an exact fast k-NN method such as this<http://research.yoonho.info/fnnne/>, which would still be much faster than a normal k-NN approach). The approximate k-NN & the exact k-NN can be used in a filter & refine method where we first obtain the X most similar image to the query image from the N images in the database using the (super fast) binary hashing approach and then re-rank these using the exact k-NN algorithm.
I would appreciate any comments/suggestions on this proposal.
Thanks,
Girish
http://www.girishmalkarnenkar.com/