Friday, March 13, 2009
Duhu (Dumb Human!)
A script that should identify a segmented letter with a certain confidence. If its confidence is above a certain value, then it should be right!
Now using this as a feedback mechanism, we try to do the segmentation and identification of just one letter. Now we decrease the segmented length and then continue detection for another letter...
Main goal is to go for a feedback based mechanism...
SVM, KNN or Neural: All the 3 seem to be similar in terms of performance. SVM and KNN cannot perform rotations unless they are trained on samples with rotation. Convolutional Neural Networks can do the rotation.
FEATURE EXTRACTION:
Calculate features using vectors. i.e. "T" means one vector to the right, the one down, and one to the right again. Store the starting position (x, y, z) and the vector (i, j, k). You can have about 5 such vectors for each character. We can scale down all vectors of a letter to some value (maybe constant height) and have an extra feature called scale for each letter. This way we can recognize letters of varying sizes. For, letters where all 5 are not utilized we can have 0s.
And, if we can rotate these vectors we will be able to make it resistant to rotation!! But how exactly does it cope with distortion and waving?
Slowing down... because of pe_dump. Will do this in my free time
Now using this as a feedback mechanism, we try to do the segmentation and identification of just one letter. Now we decrease the segmented length and then continue detection for another letter...
Main goal is to go for a feedback based mechanism...
SVM, KNN or Neural: All the 3 seem to be similar in terms of performance. SVM and KNN cannot perform rotations unless they are trained on samples with rotation. Convolutional Neural Networks can do the rotation.
FEATURE EXTRACTION:
Calculate features using vectors. i.e. "T" means one vector to the right, the one down, and one to the right again. Store the starting position (x, y, z) and the vector (i, j, k). You can have about 5 such vectors for each character. We can scale down all vectors of a letter to some value (maybe constant height) and have an extra feature called scale for each letter. This way we can recognize letters of varying sizes. For, letters where all 5 are not utilized we can have 0s.
And, if we can rotate these vectors we will be able to make it resistant to rotation!! But how exactly does it cope with distortion and waving?
Slowing down... because of pe_dump. Will do this in my free time
hi
elloh
This is the new post
I hope this will be it....
Monday, March 2, 2009
Intro....
Hi....
"Fox For All" is the plan that Uttam and I have submitted as our entry to Mozilla's "Impact Mozilla" contest. We went on to win the contest and now its time to act!!! Since the entire plan is a community based one we desperately need your comments. Please tell us what's good and more importantly what's bad about this plan. Tell us all the changes that we might make.
So, I begin with a summary of our proposal. You can get the detailed plan here:
http://docs.google.com/Doc?id=d896ncn_0cxd4sxg6
The problem statement was to improve the user retention rate of Firefox. For this, we thought of a website that will act as a platform for the entire process right from the conceptualization to the coding of addons for passive (and beginner level) users. The entire development process would be completely distributed in nature. So everything including the coding and further upgrading of add-ons will be done by the community as a
whole instead of a single person. We think that it will be very difficult for an developer to make and maintain an addon that he would never use ( He will never use it as most of these addons will have a very trivial function as far as a geek is concerned...functions like listing the names of popular websites etc...). So we thought that the
entire coding part should also be made a community work along with the idea development....
We dont expect the users themselves to give us the addon ideas...But we do expect Mozilla enthusiasts who are creatively equipped enough to think from the beginners point of view to give us add-on ideas. Once the idea is finalized we will forward the coding work to a community that will predomintantly consist of college students...These students will get an appreciation certifiate from Mozilla, mentoning the work that they have done in return for their contribution. This will be the primary motive for the students to work on the addons..Once they get their certificates another student group comes in and the work continues...
"Fox For All" is the plan that Uttam and I have submitted as our entry to Mozilla's "Impact Mozilla" contest. We went on to win the contest and now its time to act!!! Since the entire plan is a community based one we desperately need your comments. Please tell us what's good and more importantly what's bad about this plan. Tell us all the changes that we might make.
So, I begin with a summary of our proposal. You can get the detailed plan here:
http://docs.google.com/Doc?id=
The problem statement was to improve the user retention rate of Firefox. For this, we thought of a website that will act as a platform for the entire process right from the conceptualization to the coding of addons for passive (and beginner level) users. The entire development process would be completely distributed in nature. So everything including the coding and further upgrading of add-ons will be done by the community as a
whole instead of a single person. We think that it will be very difficult for an developer to make and maintain an addon that he would never use ( He will never use it as most of these addons will have a very trivial function as far as a geek is concerned...functions like listing the names of popular websites etc...). So we thought that the
entire coding part should also be made a community work along with the idea development....
We dont expect the users themselves to give us the addon ideas...But we do expect Mozilla enthusiasts who are creatively equipped enough to think from the beginners point of view to give us add-on ideas. Once the idea is finalized we will forward the coding work to a community that will predomintantly consist of college students...These students will get an appreciation certifiate from Mozilla, mentoning the work that they have done in return for their contribution. This will be the primary motive for the students to work on the addons..Once they get their certificates another student group comes in and the work continues...
As we have already talked about in our last conversation,
the entire plan was based on the firm belief that "Add-ons are not at
all being catered to the passive internet users who form a formidable
section of the worldwide internet user base. " Doing something
regarding this end, can go a long way in boosting the retention rate
of Firefox. We named our plan as "Fox for All"
We went on think of a website, something very akin to what
you have mentioned. This website will act as a platform for the entire
process of conceptualization to the coding of addons for passive (and
beginner level) users. The entire development process would be
completely distributed in nature. So everything including the coding
and further upgrading of add-ons will be done by the community as a
whole instead of a single person. We think that it will be very
difficult for an developer to make and maintain an addon that he would
never use ( He will never use it as most of these addons will have a
very trivial function as far as a geek is concerned...functions like
listing the names of popular websites etc...). So we thought that the
entire coding part should also be made a community work along with the
idea development....
We dont expect the users themselves to give us the addon
ideas...But we do expect Mozilla enthusiasts who are creatively
equipped enough to think from the beginners point of view to give us
add-on ideas. Once the idea is finalized we will forward the coding
work to a community that will predomintantly consist of college
students...These students will get an appreciation certifiate from
Mozilla, mentoning the work that they have done in return for their
contribution. This will be the primary motive for the students to work
on the addons..Once they get their certificates another student group
comes in and the work continues... This is the detailed version that
we have submitted: (Its a bit incoherent as we changed our directions
frequently while making the report)
http://docs.google.com/Doc?id=d896ncn_0cxd4sxg6
could you please tell us your thoughts about this
plan....We have a quiet a few questions to ask you with regards to the
implementation of this plan...But before I make this mail too long i
just want to end this with one of those questions....
We have so far identified two user groups that aren't being
served enough number of add-ons (Kids group on the other hand is kind
of okay with a few addons like KidZui, Glubble etc...):
1) Internet Beginners: We thought that internet beginners (I mean
those people who donot yet appreciate the full power of the
Internet)...
We can help them by showing some of the popular web sites...
We can show them how to use Web Search effectively to get almost
anything that one wants....
We can make the boring, text based Help menu ...more graphic based
and interactive
Dont you think there is a lot of scope for addons in this regard?
We also thought that a lot more of internet users fall into this
group...thus increasing the user base of such addons
2) Cyber cafe owners: This is a user group we picked up from the
spread firefox website....
Some of the addon ideas for this group are :
http://www.spreadfirefox.com/node/2861
Again this is a highly neglected group....
Please tell us of the loopholes you see in these ideas. What do
you think is the best way to get more ideas for these kind of add-ons?
We initially thought that we should set up a website for getting more
ideas but we are not sure what kind of people to expect to visit such
a website... Thanks a lot for sparing your time for us...We are highly
obliged to you
the entire plan was based on the firm belief that "Add-ons are not at
all being catered to the passive internet users who form a formidable
section of the worldwide internet user base. " Doing something
regarding this end, can go a long way in boosting the retention rate
of Firefox. We named our plan as "Fox for All"
We went on think of a website, something very akin to what
you have mentioned. This website will act as a platform for the entire
process of conceptualization to the coding of addons for passive (and
beginner level) users. The entire development process would be
completely distributed in nature. So everything including the coding
and further upgrading of add-ons will be done by the community as a
whole instead of a single person. We think that it will be very
difficult for an developer to make and maintain an addon that he would
never use ( He will never use it as most of these addons will have a
very trivial function as far as a geek is concerned...functions like
listing the names of popular websites etc...). So we thought that the
entire coding part should also be made a community work along with the
idea development....
We dont expect the users themselves to give us the addon
ideas...But we do expect Mozilla enthusiasts who are creatively
equipped enough to think from the beginners point of view to give us
add-on ideas. Once the idea is finalized we will forward the coding
work to a community that will predomintantly consist of college
students...These students will get an appreciation certifiate from
Mozilla, mentoning the work that they have done in return for their
contribution. This will be the primary motive for the students to work
on the addons..Once they get their certificates another student group
comes in and the work continues... This is the detailed version that
we have submitted: (Its a bit incoherent as we changed our directions
frequently while making the report)
http://docs.google.com/Doc?id=
could you please tell us your thoughts about this
plan....We have a quiet a few questions to ask you with regards to the
implementation of this plan...But before I make this mail too long i
just want to end this with one of those questions....
We have so far identified two user groups that aren't being
served enough number of add-ons (Kids group on the other hand is kind
of okay with a few addons like KidZui, Glubble etc...):
1) Internet Beginners: We thought that internet beginners (I mean
those people who donot yet appreciate the full power of the
Internet)...
We can help them by showing some of the popular web sites...
We can show them how to use Web Search effectively to get almost
anything that one wants....
We can make the boring, text based Help menu ...more graphic based
and interactive
Dont you think there is a lot of scope for addons in this regard?
We also thought that a lot more of internet users fall into this
group...thus increasing the user base of such addons
2) Cyber cafe owners: This is a user group we picked up from the
spread firefox website....
Some of the addon ideas for this group are :
http://www.spreadfirefox.com/
Again this is a highly neglected group....
Please tell us of the loopholes you see in these ideas. What do
you think is the best way to get more ideas for these kind of add-ons?
We initially thought that we should set up a website for getting more
ideas but we are not sure what kind of people to expect to visit such
a website... Thanks a lot for sparing your time for us...We are highly
obliged to you
Fox For All | Blogger Templates by Gecko &
Fly.
No part of the content or the blog may be reproduced without permission.
