Fwd: Ideas on UAV Tracking project

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Fwd: Ideas on UAV Tracking project

Gaurav Parida
Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems
1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.
2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]
3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

Any opinions, on which method will be better?

Additional Topics that can be handled if time permits( requires more research)
4. Handling multiple objects
5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,
Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing
[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)
     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.
PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fwd: Ideas on UAV Tracking project

tclarke
Administrator

Looks like a good start. You might want to look at the OpenCV cascade classification documentation as an alternative object detection scheme.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Wednesday, March 11, 2015 2:03 PM
To: [hidden email]
Subject: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems

1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.

2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]

3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

 

Any opinions, on which method will be better?


Additional Topics that can be handled if time permits( requires more research)

4. Handling multiple objects

5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,

Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing

[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)

     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.

PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fwd: Ideas on UAV Tracking project

Gaurav Parida
Just a quick update, I looked at Cascade Classification documentation. On looking for different algorithms, I also found the SURF algorithm[1].

As, I have my mid-semester exams till 20th March, I won't be able to come regularly on IRC and experiment with Opticks and read further for different methods for tracking. Uptil now, I have successfully installed Opticks SDK in Windows 7 with Visual Studio 2010 Ultimate and also in Ubuntu 14.04. I enlist below the todo tasks that are left to be done as soon as I am done with my midsem exams.

- Play around with python extension and experiment with the animation example.
- Learn more about the Animation framework in cpp.
- Read more documentation about plugin development.
- Do further literature survey on the chosen topic.

Thanks,
Gaurav Parida
3rd year Undergraduate Research Student,
Lab for Spatial Informatics, IIIT Hyderabad

[1] http://en.wikipedia.org/wiki/SURF

On Thu, Mar 12, 2015 at 1:25 AM, Clarke, Trevor <[hidden email]> wrote:

Looks like a good start. You might want to look at the OpenCV cascade classification documentation as an alternative object detection scheme.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Wednesday, March 11, 2015 2:03 PM
To: [hidden email]
Subject: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems

1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.

2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]

3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

 

Any opinions, on which method will be better?


Additional Topics that can be handled if time permits( requires more research)

4. Handling multiple objects

5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,

Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing

[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)

     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.

PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs



------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fwd: Ideas on UAV Tracking project

Gaurav Parida
Hello again,

Yesterday I was AFK when the meeting occurred, So I am redirecting my doubts here, that I would have asked in the meeting.

- Is there any particular format given while writing the proposal ?
- The draft proposal should be private or public, so that it can be reviewed by the mentors ?
- Any comments on the previous mail, regarding the algorithm which will be more favorable for given requirement ?

Thanks,
Gaurav Parida
3rd year Undergraduate Research Student,
Lab for Spatial Informatics, IIIT Hyderabad

On Sat, Mar 14, 2015 at 7:23 PM, Gaurav Parida <[hidden email]> wrote:
Just a quick update, I looked at Cascade Classification documentation. On looking for different algorithms, I also found the SURF algorithm[1].

As, I have my mid-semester exams till 20th March, I won't be able to come regularly on IRC and experiment with Opticks and read further for different methods for tracking. Uptil now, I have successfully installed Opticks SDK in Windows 7 with Visual Studio 2010 Ultimate and also in Ubuntu 14.04. I enlist below the todo tasks that are left to be done as soon as I am done with my midsem exams.

- Play around with python extension and experiment with the animation example.
- Learn more about the Animation framework in cpp.
- Read more documentation about plugin development.
- Do further literature survey on the chosen topic.

Thanks,
Gaurav Parida
3rd year Undergraduate Research Student,
Lab for Spatial Informatics, IIIT Hyderabad

[1] http://en.wikipedia.org/wiki/SURF

On Thu, Mar 12, 2015 at 1:25 AM, Clarke, Trevor <[hidden email]> wrote:

Looks like a good start. You might want to look at the OpenCV cascade classification documentation as an alternative object detection scheme.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Wednesday, March 11, 2015 2:03 PM
To: [hidden email]
Subject: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems

1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.

2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]

3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

 

Any opinions, on which method will be better?


Additional Topics that can be handled if time permits( requires more research)

4. Handling multiple objects

5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,

Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing

[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)

     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.

PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs




------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fwd: Ideas on UAV Tracking project

tclarke
Administrator

You need to apply via Google Melange: http://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2015/help_page#4._How_does_a_student_apply

 

In addition there are some OSGEO questions on the OSGEO GSoC page: http://wiki.osgeo.org/wiki/Google_Summer_of_Code_Recommendations_for_Students

 

I’d suggest writing up a draft and sending it for review and we’ll give you some feedback…primarily on missing information, etc.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Tuesday, March 17, 2015 1:08 AM
To: [hidden email]
Cc: Clarke, Trevor
Subject: Re: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello again,

 

Yesterday I was AFK when the meeting occurred, So I am redirecting my doubts here, that I would have asked in the meeting.

 

- Is there any particular format given while writing the proposal ?

- The draft proposal should be private or public, so that it can be reviewed by the mentors ?

- Any comments on the previous mail, regarding the algorithm which will be more favorable for given requirement ?

 

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad

Image removed by sender.

 

On Sat, Mar 14, 2015 at 7:23 PM, Gaurav Parida <[hidden email]> wrote:

Just a quick update, I looked at Cascade Classification documentation. On looking for different algorithms, I also found the SURF algorithm[1].

As, I have my mid-semester exams till 20th March, I won't be able to come regularly on IRC and experiment with Opticks and read further for different methods for tracking. Uptil now, I have successfully installed Opticks SDK in Windows 7 with Visual Studio 2010 Ultimate and also in Ubuntu 14.04. I enlist below the todo tasks that are left to be done as soon as I am done with my midsem exams.

- Play around with python extension and experiment with the animation example.

- Learn more about the Animation framework in cpp.

- Read more documentation about plugin development.

- Do further literature survey on the chosen topic.

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad


[1] http://en.wikipedia.org/wiki/SURF

On Thu, Mar 12, 2015 at 1:25 AM, Clarke, Trevor <[hidden email]> wrote:

Looks like a good start. You might want to look at the OpenCV cascade classification documentation as an alternative object detection scheme.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Wednesday, March 11, 2015 2:03 PM
To: [hidden email]
Subject: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems

1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.

2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]

3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

 

Any opinions, on which method will be better?


Additional Topics that can be handled if time permits( requires more research)

4. Handling multiple objects

5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,

Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing

[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)

     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.

PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

 

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs

 

 



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fwd: Ideas on UAV Tracking project

Gaurav Parida
Thanks for the quick reply, I will do the necessary and send the draft to you for review, as soon as I am done with my exams. 

Regards,
Gaurav Parida
3rd year Undergraduate Research Student
Lab for Spatial Informatics, IIIT Hyderabad

On Tue, Mar 17, 2015 at 7:17 PM, Clarke, Trevor <[hidden email]> wrote:

You need to apply via Google Melange: http://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2015/help_page#4._How_does_a_student_apply

 

In addition there are some OSGEO questions on the OSGEO GSoC page: http://wiki.osgeo.org/wiki/Google_Summer_of_Code_Recommendations_for_Students

 

I’d suggest writing up a draft and sending it for review and we’ll give you some feedback…primarily on missing information, etc.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Tuesday, March 17, 2015 1:08 AM
To: [hidden email]
Cc: Clarke, Trevor
Subject: Re: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello again,

 

Yesterday I was AFK when the meeting occurred, So I am redirecting my doubts here, that I would have asked in the meeting.

 

- Is there any particular format given while writing the proposal ?

- The draft proposal should be private or public, so that it can be reviewed by the mentors ?

- Any comments on the previous mail, regarding the algorithm which will be more favorable for given requirement ?

 

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad

Image removed by sender.

 

On Sat, Mar 14, 2015 at 7:23 PM, Gaurav Parida <[hidden email]> wrote:

Just a quick update, I looked at Cascade Classification documentation. On looking for different algorithms, I also found the SURF algorithm[1].

As, I have my mid-semester exams till 20th March, I won't be able to come regularly on IRC and experiment with Opticks and read further for different methods for tracking. Uptil now, I have successfully installed Opticks SDK in Windows 7 with Visual Studio 2010 Ultimate and also in Ubuntu 14.04. I enlist below the todo tasks that are left to be done as soon as I am done with my midsem exams.

- Play around with python extension and experiment with the animation example.

- Learn more about the Animation framework in cpp.

- Read more documentation about plugin development.

- Do further literature survey on the chosen topic.

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad


[1] http://en.wikipedia.org/wiki/SURF

On Thu, Mar 12, 2015 at 1:25 AM, Clarke, Trevor <[hidden email]> wrote:

Looks like a good start. You might want to look at the OpenCV cascade classification documentation as an alternative object detection scheme.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Wednesday, March 11, 2015 2:03 PM
To: [hidden email]
Subject: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems

1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.

2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]

3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

 

Any opinions, on which method will be better?


Additional Topics that can be handled if time permits( requires more research)

4. Handling multiple objects

5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,

Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing

[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)

     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.

PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

 

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs

 

 



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.


------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fwd: Ideas on UAV Tracking project

Gaurav Parida
Hello,

Sorry for the delay, I have uploaded my proposal on Google Melange as well as on Google Drive.

Please review and give your feedback to improve the proposal. Based on the suggestions, I will change the proposal submitted on Google Melange accordingly. Thank you.

Regards,
Gaurav Parida
3rd year Undergraduate Research Student
Lab for Spatial Informatics, IIIT Hyderabad

On Tue, Mar 17, 2015 at 7:21 PM, Gaurav Parida <[hidden email]> wrote:
Thanks for the quick reply, I will do the necessary and send the draft to you for review, as soon as I am done with my exams. 

Regards,
Gaurav Parida
3rd year Undergraduate Research Student
Lab for Spatial Informatics, IIIT Hyderabad

On Tue, Mar 17, 2015 at 7:17 PM, Clarke, Trevor <[hidden email]> wrote:

You need to apply via Google Melange: http://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2015/help_page#4._How_does_a_student_apply

 

In addition there are some OSGEO questions on the OSGEO GSoC page: http://wiki.osgeo.org/wiki/Google_Summer_of_Code_Recommendations_for_Students

 

I’d suggest writing up a draft and sending it for review and we’ll give you some feedback…primarily on missing information, etc.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Tuesday, March 17, 2015 1:08 AM
To: [hidden email]
Cc: Clarke, Trevor
Subject: Re: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello again,

 

Yesterday I was AFK when the meeting occurred, So I am redirecting my doubts here, that I would have asked in the meeting.

 

- Is there any particular format given while writing the proposal ?

- The draft proposal should be private or public, so that it can be reviewed by the mentors ?

- Any comments on the previous mail, regarding the algorithm which will be more favorable for given requirement ?

 

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad

Image removed by sender.

 

On Sat, Mar 14, 2015 at 7:23 PM, Gaurav Parida <[hidden email]> wrote:

Just a quick update, I looked at Cascade Classification documentation. On looking for different algorithms, I also found the SURF algorithm[1].

As, I have my mid-semester exams till 20th March, I won't be able to come regularly on IRC and experiment with Opticks and read further for different methods for tracking. Uptil now, I have successfully installed Opticks SDK in Windows 7 with Visual Studio 2010 Ultimate and also in Ubuntu 14.04. I enlist below the todo tasks that are left to be done as soon as I am done with my midsem exams.

- Play around with python extension and experiment with the animation example.

- Learn more about the Animation framework in cpp.

- Read more documentation about plugin development.

- Do further literature survey on the chosen topic.

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad


[1] http://en.wikipedia.org/wiki/SURF

On Thu, Mar 12, 2015 at 1:25 AM, Clarke, Trevor <[hidden email]> wrote:

Looks like a good start. You might want to look at the OpenCV cascade classification documentation as an alternative object detection scheme.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Wednesday, March 11, 2015 2:03 PM
To: [hidden email]
Subject: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems

1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.

2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]

3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

 

Any opinions, on which method will be better?


Additional Topics that can be handled if time permits( requires more research)

4. Handling multiple objects

5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,

Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing

[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)

     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.

PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

 

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs

 

 



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.



------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fwd: Ideas on UAV Tracking project

tclarke
Administrator

I started looking over this and I’m not quite sure you’re on the right track. I provided more specific comments in the google doc but in summary, you need to specify what data you’ll be using and what sorts of objects you are tracking. Based on this, I suspect you’re algorithm choices won’t work well. You’re also talking about reimplementing component algorithms in Opticks instead of using the OpenCV versions and creating a robust tracking system.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Tuesday, March 24, 2015 1:58 PM
To: Clarke, Trevor
Cc: [hidden email]
Subject: Re: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

Sorry for the delay, I have uploaded my proposal on Google Melange as well as on Google Drive.

 

Please review and give your feedback to improve the proposal. Based on the suggestions, I will change the proposal submitted on Google Melange accordingly. Thank you.

Regards,

Gaurav Parida

3rd year Undergraduate Research Student

Lab for Spatial Informatics, IIIT Hyderabad

 

On Tue, Mar 17, 2015 at 7:21 PM, Gaurav Parida <[hidden email]> wrote:

Thanks for the quick reply, I will do the necessary and send the draft to you for review, as soon as I am done with my exams. 

 

Regards,

Gaurav Parida

3rd year Undergraduate Research Student

Lab for Spatial Informatics, IIIT Hyderabad

 

On Tue, Mar 17, 2015 at 7:17 PM, Clarke, Trevor <[hidden email]> wrote:

You need to apply via Google Melange: http://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2015/help_page#4._How_does_a_student_apply

 

In addition there are some OSGEO questions on the OSGEO GSoC page: http://wiki.osgeo.org/wiki/Google_Summer_of_Code_Recommendations_for_Students

 

I’d suggest writing up a draft and sending it for review and we’ll give you some feedback…primarily on missing information, etc.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Tuesday, March 17, 2015 1:08 AM
To: [hidden email]
Cc: Clarke, Trevor
Subject: Re: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello again,

 

Yesterday I was AFK when the meeting occurred, So I am redirecting my doubts here, that I would have asked in the meeting.

 

- Is there any particular format given while writing the proposal ?

- The draft proposal should be private or public, so that it can be reviewed by the mentors ?

- Any comments on the previous mail, regarding the algorithm which will be more favorable for given requirement ?

 

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad

Image removed by sender.

 

On Sat, Mar 14, 2015 at 7:23 PM, Gaurav Parida <[hidden email]> wrote:

Just a quick update, I looked at Cascade Classification documentation. On looking for different algorithms, I also found the SURF algorithm[1].

As, I have my mid-semester exams till 20th March, I won't be able to come regularly on IRC and experiment with Opticks and read further for different methods for tracking. Uptil now, I have successfully installed Opticks SDK in Windows 7 with Visual Studio 2010 Ultimate and also in Ubuntu 14.04. I enlist below the todo tasks that are left to be done as soon as I am done with my midsem exams.

- Play around with python extension and experiment with the animation example.

- Learn more about the Animation framework in cpp.

- Read more documentation about plugin development.

- Do further literature survey on the chosen topic.

Thanks,

Gaurav Parida

3rd year Undergraduate Research Student,

Lab for Spatial Informatics, IIIT Hyderabad


[1] http://en.wikipedia.org/wiki/SURF

On Thu, Mar 12, 2015 at 1:25 AM, Clarke, Trevor <[hidden email]> wrote:

Looks like a good start. You might want to look at the OpenCV cascade classification documentation as an alternative object detection scheme.

 

----------------------

Trevor R.H. Clarke

Software Engineer, Ball Aerospace

(937)320-7087

 

From: Gaurav Parida [mailto:[hidden email]]
Sent: Wednesday, March 11, 2015 2:03 PM
To: [hidden email]
Subject: [Opticks-devs] Fwd: Ideas on UAV Tracking project

 

Hello,

UAV Tracking project interested me a lot and after some preliminary survey of the related literature, I would like to present some ideas about them which can be implemented for the plugin as the GSoC project. For this phase, I have read two research papers, which have been attached in the mail.

So, UAV Tracking can be divided into three subproblems

1. Object representation - A subimage/ template that has to be tracked is selected from the video frame.

2. Object Detection - Template Matching problem[1] / or storing histograms of the template for matching and then detecting it using Bhattacharya Distance[2]

3. Object Tracking - Kalman filter[1] / Condensation Algorithm[2]

 

Any opinions, on which method will be better?


Additional Topics that can be handled if time permits( requires more research)

4. Handling multiple objects

5. Handling occlusions

The ideas presented are still in research stage and not complete, as the implementation details haven't been worked upon till now. The animation framework has been worked, in this python script given in the samples[3]. I saw last year's gsoc project and it used some header files of opencv for some functions, so as of now I saw this page [4] and it correlates to the second subproblem of detection / template matching.

For the workflow that I have thought of the project as of now is, user will load the video and then play it, till the object to be detected is found for the first time and then the video will be paused and the user will select a subimage / object that has to be tracked. Now, the plugin will run object detection and then track it using the implemented techniques. The resulting video that will be generated, which will have the detected object in a rectangle. and maybe even show a trajectory of the detected object.

Any opinion and suggestion? Thanks for reading.

Regards,

Gaurav

[1] On-Board Visual Tracking with Unmanned Aircraft System (UAS). Ashraf Qadir, Jeremiah Neubert and William Semke.
     https://drive.google.com/file/d/0B4SlnIUgdUw5d0VPSlBBWTlSOWFTSm0yUFlCMFNaVk0zZ18w/view?usp=sharing

[2] Chasing a moving target from UAV. Celine Teuliere and Eric Marchand. (attached in the mail)

     https://drive.google.com/file/d/0B4SlnIUgdUw5d2FrYVRwTGlDX09wV2JSelZVb1Y3SDZpZTV3/view?usp=sharing
[3] http://opticks.org/confluence/display/opticksExtras/Using+animation+callbacks+to+stabilize+video
[4] http://docs.opencv.org/doc/tutorials/imgproc/histograms/template_matching/template_matching.html


PS : I found two thesis done on this topic, I have till now just skimmed few pages of them. As, soon as I am done with those, I will present my ideas from them to the community.

PPS: Sorry for any grammatical errors and typos in the mail. English is not my first language.



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

 

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs

 

 



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

 

 



This message and any enclosures are intended only for the addressee. Please
notify the sender by email if you are not the intended recipient. If you are
not the intended recipient, you may not use, copy, disclose, or distribute this
message or its contents or enclosures to any other person and any such actions
may be unlawful. Ball reserves the right to monitor and review all messages
and enclosures sent to or from this email address.

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Opticks-devs mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/opticks-devs
Loading...