Graduate School and Research Center in Digital Sciences

IRIM at TRECVID 2013: Semantic indexing and multimedia instance search

Ballas, Nicolas et al.

TRECVID 2013, International Workshop on Video Retrieval Evaluation, 20-22 November 2013, Gaithersburg, MD, USA

The IRIM group is a consortium of French teams working on Multimedia Indexing and Retrieval. This paper describes its participation to the TRECVID 2013 semantic indexing and instance search tasks. For the semantic indexing task, our approach uses a six-stages processing pipelines for computing scores for the likelihood of a video shot to contain a target concept. These scores are then used for producing a ranked list of images or shots that are the most likely to contain the target concept. The pipeline is composed of the following steps: descriptor extraction, descriptor optimization, classi cation, fusion of descriptor variants, higher-level fusion, and re-ranking. We evaluated a number of dif-ferent descriptors and tried di erent fusion strategies. The best IRIM run has a Mean Inferred Average Precision of 0.2796, which ranked us 4th out of 26 participants.

Document Bibtex

Title:IRIM at TRECVID 2013: Semantic indexing and multimedia instance search
Type:Conference
Language:English
City:Gaithersburg
Country:UNITED STATES
Date:
Department:Data Science
Eurecom ref:4729
Copyright: © NIST. Personal use of this material is permitted. The definitive version of this paper was published in TRECVID 2013, International Workshop on Video Retrieval Evaluation, 20-22 November 2013, Gaithersburg, MD, USA and is available at :
Bibtex: @inproceedings{EURECOM+4729, year = {2013}, title = {{IRIM} at {TRECVID} 2013: {S}emantic indexing and multimedia instance search}, author = {{B}allas, {N}icolas et al. }, booktitle = {{TRECVID} 2013, {I}nternational {W}orkshop on {V}ideo {R}etrieval {E}valuation, 20-22 {N}ovember 2013, {G}aithersburg, {MD}, {USA}}, address = {{G}aithersburg, {UNITED} {STATES}}, month = {11}, url = {http://www.eurecom.fr/publication/4729} }
See also: