Re: Is there anynone who knows a chatting site or someting like this? - Face Matching



On Mar 9, 3:15 am, dondora <koni...@xxxxxxxxxxx> wrote:
You understood what I wrote.
I think this project is related to face recognition and searching for
how similar a normal person is compared to someone posted by a user
who is a user's ideal type.
I've got an idea to provide userlist containing user's contact
information
whose similarity degree evaluated by the system  is greater than 50
or
something(assuming 100 is full score which means almost the person in
the list
and a person in the photo posted by the user are the same and 0 means
never cannot find any similarity)

I think there's a need to recognize face and analyze the face to
compare
all the user's photos in database properly designed and express
numericaly how similar one is.

Those are things in my mind. Anyway now could you tell me where I can
get libraries to implement this system?
This is my school term project and my professor wants all students
have new idea that doesn't exist now.
Please help me~!~!~!~!
----------------------------------
dondora:
Well then, as far as I know, this would be an idea that doesn't exist
now -- probably because there's no need, calling, funding, or market
for this type of service yet. At least not as a regular business, but
there may be usage based on the fun/curiosity/entertainment aspect of
it like there is for such web sites as http://www.hotornot.com/. But
ignoring whether any web site could make money at this, you could
continue with your project describing how users could upload a photo
of a face, and then you would search your database to find matching
faces. (You should have used "face recognition" or "face matching" in
your original subject line). You would build up your database in
advance by searching (Google-images-like) the internet retrieving
faces and processing the face for certain characteristics (like hair
color, shape, etc.) to get a feature vector. Then store the face's
URL and the feature vector to your database. You would compare the
feature vector of the uploaded face to your database to retrieve
possible similar faces. Maybe the user could even change something
like saying they want "matched" faces to have a certain hair or eye
color, or sex. (Might be hard to determine sex of a face, randomly
retrieved from the internet, automatically though.)

Now do a search of face matching to get some more ideas and you're on
your way to finishing your project.
Regards,
ImageAnalyst
.



Relevant Pages

  • Re: File Auditing - Fast DB import and data manipulation
    ... the files into the database is not the best way forward. ... on either system (you'll need to download gunzip from gnuwin32 ... The databases I've dealt with (Oracle, SQL Server, Access, Postgres) ... So, I need some way to compare todays file, to yesterdays and see ...
    (comp.lang.ruby)
  • RE: Database running over network runs extremely slow
    ... I have reports built into the database that track the entry ... you are searching (or using in your DLOOKUPs criteria parameter)? ... of the DLookups to looped recordset searches, ...
    (microsoft.public.access.tablesdbdesign)
  • Re: rules engine ideas? Trying to prevent tons of conditional branches in a logic filter.
    ... One thing you could do is to place your rules in a database. ... consists of an XPath expression for the operand, and operator, the value you ... compare against, and the score. ... apply each of the Xpath queries to ...
    (microsoft.public.dotnet.languages.csharp)
  • Re: Divining the full pathname of a file, all logicals translated
    ... compare it to the "current" state when an audit report is run. ... drive name needs to be retained also, unmasked from any logicals. ... The auditors may never see the actual database. ...
    (comp.os.vms)
  • Re: Efficient Integer Comparison Algorithm
    ... The other database is external and I ... Do you think there is a way to link my Oracle DB and the file input ... Since you only want to compare according to equality of your two values a HashMap is more efficient. ... If the first ordered column in the local store data is less than in the DB data, increase it until they are same or greater. ...
    (comp.lang.java.programmer)