logo
upload images photo quiz image analysis blog about
 
 

Technology Used:

The See What You Feel site is written in HTML with external CSS style sheets and embedded Flash animations. The Flash interactions are coded in ActionScript, along with XML connector calls. Going forward, all image galleries will be built using PHP and XML calls for content display. The XML will be dynamically generated through XSLT and PHP server-side script. Currently the Flash galleries are generated with static XML pointers pulling images from a file directory.

During our research on possible image analysis applications we could use, we were extremely fortunate to discover an open source application, isk-daemon (http://server.imgseek.net/), developed by Ricardo Cabral.

Isk-daemon allows content based image retrieval of any photo website or database. Cabral applied the wavelet algorithms by Alfred Haar along with metric and query ideas that were based on the work of Jacobs, Finkelstein and Salesin (1995).

We are currently working on how to query images from isk-daemon and integrate it into our front-end. Additionally, we continue to research other possible backend solutions for web-based image search and analysis.

We are using an off the shelf PHP and MySQL photo application (open source) to manage images and to interact with the image analysis application. The photo gallery includes GD image library functionality which will allow us to tailor the size of photographs dynamically.

Interested in being a part of this project?
We are seeking a programmer who can provide technical advice and feedback on XML_RPC, XSLT framework, Flash (ActionScript), PHP, Python and MySQL. While we’ll be doing the bulk of the coding and building of the website, we are looking for someone who’s interested in the project and who would be interested in giving us feedback and guidance when we encounter technical issues or questions.

We are also seeking someone who is familiar with image analysis and can provide direction and guidance as we work on identifying images by color and composition.

To contact us, email group@seewhatyoufeel.com.

 

References
Fast Multiresolution Image Querying. Jacobs, Charles E., Adam Finkelstein, and David H. Salesin. In Proceedings of SIGGRAPH '95, pages 277-286. ACM, New York, 1995. http://grail.cs.washington.edu/projects/query/