Recently Savvas Chatzichristofis released a Facebook app based on his tool img(Rummager) to retrieve the photos of Facebook contacts based on image content. The tool – called img(Finder) – allows to connect to a Facebook account and creates an index based on the CEDD and FCTH image features. img(Finder) indexes the images of all Facebook friends. Querying the index is done by sketch (painting tool is included) or by sample image. Best of all: The tool is open source
Emptying the TF card of my phone I finally uploaded some photos from the final presentations to Flickr. It’s just five from my camera phone, but I hope to get another batch from a colleague.
Recently we got accepted at the open source contest at the ACM Multimedia this year in Vancouver. I finished our contributions few days ago and here I’m sharing a draft version of it. Most interesting in my opinion are the the performance figures for LIRe. Obviously CEDD performs very well in terms of runtime performance and subjective query results.
Mathias Lux & Savvas A. Chatzichristofis, “LIRe: Lucene Image Retrieval – An Extensible Java CBIR Library”, in Proceedings ACM MM 2008, Vancouver, CA, Nov. 2008 (to appear)
I uploaded paper and presentation for the KASW’08 workshop: Mathias Lux, Oge Marques & Arthur Pitman, “Using Visual Features to Improve Tag Suggestions in Image Sharing Sites”, in Proceedings Workshop on Knowledge Acquisition from the Social Web, Graz, Austria, Sept. 2008
The rest of the Triple-I was straight forward for me. I did a lot of talking with old colleages and industry people interested in multimedia asset management. Looks like everybody collected a huge amount of digital photos and is now searching for means to organize them
Next years Triple-I will be somewhat different. Having found a call for a next year’s I-Know I think the Triple-I will go back to the original I-Know as it seems that especially the concept of the I-semantics didn’t work out. The next event of the Multimedia Metadata Community will be our 9th workshop in Toulouse.
The first day here at the Triple-I was very successful: The KASW ’08 workshop went great, presentations were insightful and discussions were lively. My presentation was also great although I could have talked longer (the talk scheduled after mine did not take place due to visa issues). Highlights were definitely the overview of research on folksonomies of Andreas Hotho, Ralf Klamma’s summary of the finding of his group regarding online communites (see Flickr for a quote) and the position of Andrew Gordon regarding the blogosphere and common sense knowledge.
Today I already chaired the special session of the multimedia metadata community at the I-Media with three interesting presentations. Besides all the scientific stuff I had a lot of chats about industry needs, computer games on an academic level, my eee pc and my trip to the GC
Tomorrow morning I’m on the road to go to Graz, where the Triple-I (former I-Know) takes place Wednesday to Friday. I’ll be there to visit and present at the KASW workshop and to chair the session of the multimedia metadata community. I’m looking forward to meet my colleages there. I-Know is always productive and good fun.