Job offer : engineer position in mutlimedia retrieval systems' evaluation

Position: research engineer.
Title: evaluation of content-based image and video document indexing systems
Duration: from 18 to 24 months.
Target starting date: 1st November 2011.
Location: Laboratory of Informatics of Grenoble: http://www.liglab.fr/
Team: Multimedia Information Indexing and Retrieval: http://mrim.imag.fr/
Salary: between 1900 and 2200 € net per month depending upon experience.
Contact: Georges Quénot (Researcher at CNRS, HDR), Georges.Quenot@imag.fr.

In the context of the Quaero Programme (http://www.quaero.org), the recruited person will:
* participate to the development or adaptation of image and video corpus annotation tools;
* manage the use of these tools by a team of annotators for the effective creation of annotated corpus;
* participate to the creation or adaptation of tools for the evaluation of content-based image and video document indexing systems;
* participate to the organization of evaluation campaigns for such systems;
* participate to the administrative management of the project.
Expected skills: Unix/Linux, Windows, C/C++, Java, XML, HTML/CGI.
http://mrim.imag.fr/en/Positions/poste-IQ1b.pdf

Congratulations to DEMIR research team for winning the ImageCLEF-MED 2011 Ad-hoc image-based retrieval task using CEDD

Congratulations to DEMIR research team for winning the ImageCLEF-MED 2011 Ad-hoc image-based retrieval task using CEDD!!!!

Abstract. This paper present the details of participation of DEMIR  (Dokuz Eylul University Multimedia Information  Retrieval) research team to the context of our participation to the ImageCLEF 2011 Medical Retrieval task.  This year, we evaluated fusion and re-ranking method which is based on the best low level feature of images with best text retrieval result. We improved results by examination of different weighting models for retrieved text data and low level features. We tested multi–modality image retrieval in ImageCLEF 2011 medical retrieval task and obtained the best seven ranks in mixed retrieval, which includes textual and visual modalities. The results clearly show that proper fusion of different modalities improve the overall retrieval performance.

Read the paper http://clef2011.org/resources/proceedings/Alpkocak_Clef2011.pdf