Video Processing Help Needed

Question

1.00/5 (2 votes)

See more:

Hello Friends,

Currently i am working on project which analyses two videos.
I want to compare how similar action you can perform as compared to other
Suppose there is video showing how to dance n other one is you doing same dance
So i have to compare them !!!!
Plz guide me :)

Posted 5-Oct-11 3:28am

Sumeet.kate

Add a Solution

1 solution

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Simon Bang Terkildsen · Answer 1 · 2011-10-05T03:51:00

Solution 1

So i have to compare them
Come on this is something that is not even close to be perfected; Way to big a question and impossible to answer completely.

I would suggest using the Kinect SDK[^] to work with skeletons and then compare the movement of the skeleton of the user with expected movement of the base skeleton.

This however is not easy it would be worth a PhD thesis, and since you asked this question I'm pretty sure you should forget about it and move on, as it's beyond you at the moment.
This might sound harsh, but you don't learn anything by taking on a project you're doomed to fail, you'll only waste your time and knock down your confidence more or less.

Posted 5-Oct-11 3:51am

Simon Bang Terkildsen

Comments

Simon Bang Terkildsen 5-Oct-11 10:09am

Well I suppose I could give you some projects you might be able to do (though hard as I can't really judge your current level precisely)
If you really want to just code, as in do a project, I suggest you look through the article archive of the code project and find something you find interesting and create that yourself, that way you have examples and hopefully some explanations in the article. However for you to learn anything from doing that you have to be very critical about the code and design of the code, thus you would do a lot of googling to find alternative solutions to a specific piece of code and look ups in documentations.

The best way I think you can learn is by getting a good book, read it and do all the exercises.

Sumeet.kate 5-Oct-11 10:03am

Hey do you have any project ideas
Fairly complex but possible to implement ??

Simon Bang Terkildsen 5-Oct-11 10:10am

See my comment to your other comment which has disappeared.

Sumeet.kate 5-Oct-11 10:15am

Hey thank for showing interest :)
Hey i am final year engg student !!!!
Basically need project idea for my final year also need to publish paper!!!!

Simon Bang Terkildsen 5-Oct-11 10:19am

Ah ok, again I could just suggest some projects, but I don't really know what you're capable of and most importantly what your interests are.
A lot of similar questions has been posted in the past,

Sumeet.kate 5-Oct-11 10:50am

Okies :)

But still if you hav any ideas let me noe :)

Espen Harlinn 5-Oct-11 10:42am

Good reply :)

Sergey Alexandrovich Kryukov 5-Oct-11 12:49pm

PhD thesis? Are you kidding?! The most of even the best dissertations are not even close to this level. The problem is way more serious. Solution is nearly impossible (but first of all because criteria of "closeness" are completely undefined). I don't even think modern level of recognition science and technology is close enough to compare something like dance. Whole words is still messing with body gesture recognition, where there is quite a reasonable progress, but I think this task is way more simple. Some really innovative approach to relatively simple gesture recognition could make a good dissertation.

This is my own judgement though, I could be wrong.

The answer is generally good, especially the follow-up discussion; I voted 5.
--SA

Simon Bang Terkildsen 6-Oct-11 11:55am

Thank you for the vote.
Well a PhD thesis doesn't have to take on the entire problem or even provide a general solution to a problem. So while you're correct taking on the OP's problem directly without any prior research to rely on, I still feel that it would be worth a PhD thesis. Now I'm not a researcher but I would start by using the research you mentioned about body gesture recognition, So you would look for the correct gestures in the correct order, once that is done I would go deeper.

Sergey Alexandrovich Kryukov 25-Oct-11 22:56pm

Well... maybe. I just want to note that gesture recognition is way more simple. As to comparison of two videos, the closeness in simply not defined, to start with. This is not a trivial issue. I remember several years ago a technology was demonstrated to search for a painting by the "picture" sketched by the user. I cannot say it was impressive...
--SA