Hi Sameer, Thanks for sending along the info about your code projects. I don't use a Windows machine so it is a bit more difficult for me to run the exe files. Jon will be taking a look at them later though. I was able to look at the lightbulb intro page, so that's good.
Looking forward to seeing your proposal. At the moment the only thing that sticks out is that you may have not thought through the keyboard accessibility fully. This will likey be dependent on both your overall design and which features you plan to implement. You don't need to be able to solve all of the issues right now, but make sure you schedule time in to think about these. Also, there has been a lot of interest in this project, you might want to apply to some others as well, to increase your chances of landing a project. Thanks Justin On 2012-04-02, at 10:57 AM, Sameer Goyal wrote: > Hi, > > Glad to hear from you. Regarding the features you were talking about, the > undo/redo operation should be pretty easy to implement by keeping a simple > stack of the operations that are being done on the image. Also, making the > interface keyboard accessible should also be pretty straightforward since we > just need to attach the same event handler to the keys as well as the click > on the respective tool in the toolbox. Regarding automatic deskewing of the > image, we can provide a tool like straighten so that the user can choose how > much should the image be sheared to straighten it. There are algorithms that > can automate this process, but the accuracy of the algorithms depends hugely > on the quality of the image & the contrast between the actual image & the > background. Anyways, I am still looking into it as of now. I understand that > I'd be expected to make use of Fluid Infusion, and since it is built on top > of jQuery, I should be able to work on it without much difficulty. I am going > through the documentation at the moment. > > Basically, bits-apogee (2011) was developed by a team of three people > including a designer, a front end developer & a back end > developer(http://www.bits-apogee.org/2011/webteam/). I was the front end > developer, so my tasks included converting first the given PSD to HTML, & > then adding functionalities for displaying information regarding various > events & also providing an interface for the user to manage his events, > teams, registration status, etc in the my profile tab. There was a large > usage of AJAX requests to a Django server to get the data for the user. In > 2012, I was working alongside another front end developer & for providing > unique URLs to each event page we made use of the HTML5 history API. Earlier > we were doing it using hash tags. > > Also, we made an intro page which was there on the site before the complete > site was prepared. http://www.bits-apogee.org/intro/ > > I had made use of the Pixastic library to manipulate the image of a bulb > depending on the mouse position in the browser window. If you move the mouse > on the y axis, you will see the hue of the bulb getting manipulated & if you > move your mouse on the x axis, you will see the saturation is maximum at a > point just where there is apogee text & drops down to 0 at both ends. > > I am attaching some of my image processing work as well, there is an > implementation of a median filter to remove noise from an input image, & an > implementation of brightness adjust which can enhance the brightness as well > as reduce it. Currently the only allowed input format is ppm since it > provides the pixel matrix directly which we can easily get in javascript > through canvas functions for any format. I will be writing my proposal on the > GSoC site within a day or two. Looking forward to more discussions with you & > Jon. > Regards, > > Sameer Goyal > > B.E.(Hons.) Computer Science > > BITS Pilani, India > > +91-8058426750 > > [email protected] > > On Sun, Apr 1, 2012 at 9:04 PM, Justin Obara <[email protected]> wrote: > Hello Sameer, > > Thanks for your interest in the HTML5 Image Editor. I've cc'd Jon on this > e-mail. He's a designer here who will be helping to mentor this project. I'm > sure both of us would like to hear more about your ideas. Some additional > features/areas we'd like to have incorporated would be undo/redo, automatic > deskewing of the image, and, at the least, making the interface keyboard > accessible. In terms of external libraries, while it should generally be okay > to make use of them, we'd have to determine it on a case-by-case basis. > Additionally, we would expect you to make use of Fluid Infusion to implement > the code. > > I wasn't quite clear as to what you have done at bits-apogee. Could you > please clearly describe how you were involved there? Also, examples > demonstrating your work would be useful. > > Here are some useful links. > > IRC Channel info: > http://wiki.fluidproject.org/display/fluid/IRC+Channel > > Infusion Documentation: > http://wiki.fluidproject.org/display/docs/Infusion+Documentation > > Organization GSoC page (application template at the bottom): > http://www.google-melange.com/gsoc/org/google/gsoc2012/idi_idrc > > GSoC Student guide: > http://en.flossmanuals.net/GSoCStudentGuide/ > > Thanks > Justin > > On 2012-03-31, at 7:24 PM, Sameer Goyal wrote: > >> >> Hi, >> I am Sameer Goyal, a third year student of BITS, Pilani, India. I have a >> keen interest in pursuing one of your projects - HTML5 Image Editor as part >> of the GSoC 2012 program. >> >> I have been a front end developer for 3 years, & an avid follower of the new >> developments in the current HTML5 standard. I have strong skills in >> Javascript, HTML5 canvas, HTML5 video, CSS3. I am quite familiar with canvas >> & various APIs that go along with it. I have also covered an elective course >> Image Processing & am well equipped with various image manipulation >> techniques & how to work with pixel data to create various popular effects. >> Some of my previous works include - www.bits-apogee.org/2012, >> www.bits-apogee.org/2011. >> >> I have gone through the project details, and I have some additional ideas >> that can be implemented to make the project better & well rounded. >> Since brightness/contrast & thresholding operations are being implemented, >> we might consider using image manipulation libraries that work with the >> ImageData API provided along with canvas. Eg - Pixastic >> (http://www.pixastic.com/) & a variety of other alternatives exist. >> Even if we choose to avoid using external libraries, we can implement a >> variety of other adjustments by simply working with the ImageData. >> Layers - We can implement layers, that is inherent in all modern image >> editors. Also, blend modes can be implemented between layers. (Although, >> there maybe a performance issue for cases where the images are large in size) >> Masks - We can also provide support for masking using canvas' compositing >> operations. >> We can provide a toolbox containing tools for various transformations >> (rotate, skew, scale) so that the interface is highly intuitive. >> Looking forward to hear from you about the project. >> >> Regards, >> >> Sameer Goyal >> >> B.E.(Hons.) Computer Science >> >> BITS Pilani, India >> >> +91-8058426750 >> >> [email protected] >> >> _______________________________________________________ >> fluid-work mailing list - [email protected] >> To unsubscribe, change settings or access archives, >> see http://lists.idrc.ocad.ca/mailman/listinfo/fluid-work > > > <brightness.rar><median.rar>
_______________________________________________________ fluid-work mailing list - [email protected] To unsubscribe, change settings or access archives, see http://lists.idrc.ocad.ca/mailman/listinfo/fluid-work
