Friday, October 29, 2010

Unlogo - The Corporate Identity Media Filter

Unlogo is a web service, FFMPEG plugin, and AfterEffects plugin that eliminates logos and other corporate signage from videos. On a practical level, it takes back your personal media from the corporations and advertisers. On a technical level, it is a really cool combination of some brand new OpenCV and FFMPEG functionality. On a poetic level, it is a tool for focusing on what is important in the record of your life rather than the ubiquitous messages that advertisers want you to focus on.

 

In short, Unlogo gives people the opportunity to opt out of having corporate messages permanently imprinted into the photographic record of their lives.

It's simple: you upload a video, and after your video is processed, you get an email with a link to the unlogofied version (example: http://vimeo.com/14531292). You can choose to have the logos simply blocked out with a solid color or replaced with other images, such as the disembodied head of the CEO of the company. This scheme is a bit ridiculous (which is kind of my style), but I like it because it literalizes the intrusion into the record of your life that these logos represent.

What is the money for?

The majority of the money will allow me to work on Unlogo for another 2 months. These is enough time to get the filter out of training mode and into active recognition mode.

Some will go towards paying off some of the debt from the money that I have already spent on programming help and server costs.

Some of the money (probably $1k-$1.5k) will be used to pay the consultants and additional programmers who have been helping to get the filter finished and running smoothly.

Some of the money ($300-$400) will go to the hoisting bills. The CPU power to run this filter publicly isn't cheap! I have 1 Amazon EC2 server where the filtering/encoding is done, another S3 bucket where the videos are stored, and a 3rd server to host the front-end.

http://www.kickstarter.com/projects/816924031/unlogo-the-corporate-media-filter-0

Friday, October 22, 2010

Best Paper Award–2nd in a Week!!!!

My paper with Yiannis Boutalis and Avi Arampatzis

ACCELERATING IMAGE RETRIEVAL USING BINARY HAAR WAVELET TRANSFORM ON THE COLOR AND EDGE DIRECTIVITY DESCRIPTOR"

awarded as the BEST PAPER* at the “The Fifth International Multi-Conference on Computing in the Global Information Technology (ICCGI 2010)»,



Full Citation:

S. A. Chatzichristofis, Y.S. Boutalis and Avi Arampatzis, “ACCELERATING IMAGE RETRIEVAL USING BINARY HAAR WAVELET TRANSFORM ON THE COLOR AND EDGE DIRECTIVITY DESCRIPTOR”, «The Fifth International Multi-Conference on Computing in the Global Information Technology (ICCGI 2010)», Proceedings: IEEE Computer Society, pp. 41-47,  September 20 to 25, 2010, Valencia, Spain

Abstracts:

In this paper, a new accelerating technique for content-based image retrieval is proposed, suitable for the Color and Edge Directivity Descriptor (CEDD). To date, the experimental results presented in the literature have shown that the CEDD demonstrates high rates of successful retrieval in benchmark image databases. Although its storage requirements are minimal, only 54 bytes per image, the time required for the retrieval procedure may be practically too long when searching on large databases. The proposed technique utilizes the Binary Haar Wavelet Transform in order to extract from the CEDD a smaller and more efficient descriptor, with a size of less than 2 bytes per image, speeding up retrieval from large image databases. This descriptor describes the CEDD, but not necessarily the image from which is extracted. The effectiveness of the proposed method is demonstrated through experiments performed with a known benchmark database.

*9 papers were  awarded

http://www.iaria.org/conferences2010/AwardsICCGI10.html

Saturday, October 16, 2010

Best Paper Award

My paper with Vicky Kalogeiton, Dim Papadopoulos and Yiannis Boutalis

“A NOVEL VIDEO SUMMARIZATION METHOD BASED ON COMPACT COMPOSITE DESCRIPTORS AND FUZZY CLASSIFIER"

awarded as the BEST PAPER* at the “1st International Conference for Undergraduate and Postgraduate Students in Computer Engineering, Informatics, related Technologies and Applications”, October 14 to 15, 2010, Patra, Greece

zeta

Full Citation:

V. S. Kalogeiton, D. P. Papadopoulos, S. A. Chatzichristofis and Y. S. Boutalis, “A NOVEL VIDEO SUMMARIZATION METHOD BASED ON COMPACT COMPOSITE DESCRIPTORS AND FUZZY CLASSIFIER”, «1st International Conference for Undergraduate and Postgraduate Students in Computer Engineering, Informatics, related Technologies and Applications (EUREKA 2010)», October 14 to 15, 2010, Patra, Greece,

Abstract:

In this paper, a novel method to generate video summaries is proposed, which is allocated mainly for being applied to on-line videos. The novelty of this approach lies in the fact that the
authors of this paper transfer the video summarization problem to a single query image retrieval problem. This approach utilizes the recently proposed Compact Composite Descriptors (CCDs) and a fuzzy classifier. In particular, all the video frames are initially sorted according to the distance between an artificially generated, video depended, image. Then the ranking list is classified into a preset number of clusters using the Gustafson Kessel fuzzy classifier. The video abstract is calculated by extracting a representative key frame from every cluster. A significant characteristic of the proposed method is its ability to classify the frames of the video into one or more clusters. Experimental results are presented to indicate the effectiveness of the proposed approach.

Please note that both Vicky Kalogeiton and Dim Papadopoulos are undergraduate students and members of the DUTH EECE Robotics team. I feel totally honored, thank you :)

On the occasion of the award, I would like to congratulate all members of the DUTH EECE Robotics team for the high quality papers presented at the EUREKA 2010 conference.

*4 papers were  awarded

Thursday, October 14, 2010

Limited-time FREE access: International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI)

This journal publishes both applications and theory-oriented articles on new developments in the fields of pattern recognition and artificial intelligence, and is of interest to both researchers in industry and academia. From the beginning, there has always been a close relationship between the disciplines of pattern recognition and artificial intelligence. The recognition and understanding of sensory data like speech or images, which are major concerns in pattern recognition, have always been considered as important subfields of artificial intelligence.

As part of our contribution to the dissemination of scientific knowledge in the fields of Pattern Recognition and Artificial Intelligence, we have decided to extend a free access for the latest Volume of the journal. So sign up here and gain Free Access to the 2010 full text articles today!

Saturday, October 9, 2010

Greece: Funding Opportunities for Postdoctoral Research

The Greek Ministry of Education, Lifelong Learning and Religious Affairs invites all interested beneficiaries to submit proposal summaries for their inclusion in the Action "Support of Postdoctoral Researchers".

The aim of this Action is to facilitate the acquirement of new research skills by Postdoctoral Researchers (PR) that will promote their career development in any field and/or help them restart their careers after a leave of absence (but no more than seven (7) years following their doctorate conferment date). Emphasis will be given to the support of new scientists at the beginning of their career. The duration of the research projects should range from 24 to 36 months. Each candidate can submit only one (1) research proposal. The total pubic cost of the present call is 30.000.000€ and it is co-financed by the ESF (European Social Fund). The maximum budget for each project is 150.000€. At least 60% of the total budget should be related to costs pertaining to activities undertaken by the PR and should include a monthly net allowance of 1.600€, subject to cost-of-living adjustments for different countries.

Read More or Direct contact me at savvash@gmail.com

Thursday, October 7, 2010

Final Call for Papers - ECIR 2011

3rd European Conference on Information Retrieval
Dublin, Ireland
18 - 21 April 2011
In cooperation with: BCS-IRSG, Dublin City University, University of Sheffield
http://ecir2011.dcu.ie

The 33rd European Conference on Information Retrieval (ECIR 2011) will take place in Dublin, Ireland from 18-21 April 2011. ECIR provides an opportunity for both new and established researchers to present research papers reporting new, unpublished, and innovative research results within Information Retrieval. ECIR has traditionally had a strong student focus and papers whose sole or main author is a postgraduate student or postdoctoral researcher are especially welcome. As an added incentive at ECIR2011, accepted full-papers whose first author is a student will be given the opportunity to be paired with a senior mentor who will interact with them during the course of the conference. Each mentor will be an expert in the relevant area of the student's work, and will provide questions and subsequent feedback after their presentation.
We are seeking the submission of high-quality and original full-length research papers, short-papers, posters and demos, which will be reviewed by experts on the basis of the originality of the work, the validity of the results, chosen methodology, writing quality and the overall contribution to the field of Information Retrieval. Papers that demonstrate a high level of research adventure or which break out of the traditional IR paradigms are particularly welcome.
Relevant topics include, but are not limited to:

-       Enterprise Search, Intranet Search, Desktop Search, Adversarial IR
-       Web IR and Web log analysis
-       Multimedia IR
-       Digital libraries
-       IR Theory and Formal Models
-       Distributed IR, Peer-to-peer IR,
-       Mobile IR, Fusion/Combination
-       Cross-language retrieval, Multilingual retrieval, Machine translation for IR
-       Topic detection and tracking, Routing
-       Content-based filtering, Collaborative filtering, Agents, Spam filtering
-       Question answering, NLP for IR
-       Summarization, Lexical acquisition
-       Text Data Mining
-       Text Categorization, Clustering
-       Performance, Scalability, Architectures, Efficiency, Platforms
-       Indexing, Query representation, Query reformulation,
-       Structure-based representation, XML Retrieval
-       Metadata, Social networking/tagging
-       Evaluation methods and metrics, Experimental design, Test collections
-       Interactive IR, User studies, User models, Task-based IR
-       User interfaces and visualization
-       Opinion mining, Sentiment Analysis
-       Blog and online-community search
-       Other domain-specific IR (e.g., Genomic IR, legal IR, IR for chemical structures)

The ECIR 2011 conference proceedings will be published by Springer in the Lecture Notes in Computer Science (LNCS) series. Authors for both long and short papers are invited to submit their paper on or before 15 October 2010. All paper submissions must be written in English following the LNCS author guidelines [http://www.springer.com/computer/lncs?SGWID=0-164-6-793341-0]. Full papers must not exceed 12 pages including references and figures, short-papers submissions should have a maximum length of 6 pages, and posters and demos must not be longer than 4 pages. All papers will be refereed through double-blind peer review so authors should take reasonable care not identify themselves in their submissions. The proceedings will be distributed to all delegates at the Conference.

Wednesday, October 6, 2010

CBMI 2010

Following the eight successful previous events of CBMI (Toulouse 1999, Brescia 2001, Rennes 2003, Riga 2005, Bordeaux 2007, London 2008, Chania 2009 and Grenoble 2010), the Video Processing and Understanding Lab (VPULab) and the Information Retrieval Group (IRG) at Universidad Autónoma de Madrid will organize the next CBMI event.

CBMI 2011 aims at bringing together the various communities involved in the different aspects of content-based multimedia indexing, retrieval, browsing and presentation. The scientific program of CBMI 2011 will include invited keynote talks and regular and special sessions with contributed research papers.

Topics of interest, grouped in technical tracks, include, but are not limited to:

   Visual Indexing

  • Visual indexing (image, video, graphics)
  • Visual content extraction
  • Identification and tracking of semantic regions
  • Identification of semantic events

   Audio and Multi-modal Indexing

  • Audio indexing (audio, speech, music)
  • Audio content extraction
  • Multi-modal and cross-modal indexing
  • Metadata generation, coding and transformation

   Multimedia Information Retrieval

  • Multimedia retrieval (image, audio, video, …)
  • Matching and similarity search
  • Content-based search
  • Multimedia data mining
  • Multimedia recommendation
  • Large scale multimedia database management

   Multimedia Browsing and Presentation

  • Summarisation, browsing and organization of multimedia content
  • Personalization and content adaptation
  • User interaction and relevance feedback
  • Multimedia interfaces, presentation and visualization tools

Selected papers will appear, after extension and peer-review, in a special issue of Multimedia Tools and Applications.

Sunday, October 3, 2010

Co-Author Path


http://academic.research.microsoft.com/