Graduate School and Research Center in Digital Sciences

Perplexity-based evidential neural network classifier fusion using MPEG-7 low-level visual features

Benmokhtar, Rachid;Huet, Benoit

MIR 2008, ACM International Conference on Multimedia Information Retrieval 2008, October 27-November 01, 2008, Vancouver, BC, Canada

In this paper, an automatic content-based video shot index- ing framework is proposed employing five types of MPEG-7 low-level visual features (color, texture, shape, motion and face). Once the set of features representing the video content is determined, the question of how to combine their individ- ual classifier outputs according to each feature to form a final semantic decision of the shot must be addressed, in the goal of bridging the semantic gap between the low level visual feature and the high level semantic concepts. For this aim, a novel approach called "perplexity-based weighted descrip- tors" is proposed before applying our evidential combiner NNET [3], to obtain an adaptive classifier fusion PENN (Perplexity-based Evidential Neural Network). The experi- mental results conducted in the framework of the TRECVid'07 high level features extraction task report the efficiency and the improvement provided by the proposed scheme.

Document Doi Bibtex

Title:Perplexity-based evidential neural network classifier fusion using MPEG-7 low-level visual features
Keywords: classifier fusion, entropy, evidence theory, neural network, perplexity, video semantic analysis, visual descriptors
Type:Conference
Language:English
City:Vancouver
Country:CANADA
Date:
Department:Data Science
Eurecom ref:2570
Copyright: © ACM, 2008. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in MIR 2008, ACM International Conference on Multimedia Information Retrieval 2008, October 27-November 01, 2008, Vancouver, BC, Canada http://doi.acm.org/10.1145/1460096.1460151
Bibtex: @inproceedings{EURECOM+2570, doi = {http://doi.acm.org/10.1145/1460096.1460151}, year = {2008}, title = {{P}erplexity-based evidential neural network classifier fusion using {MPEG}-7 low-level visual features}, author = {{B}enmokhtar, {R}achid and {H}uet, {B}enoit}, booktitle = {{MIR} 2008, {ACM} {I}nternational {C}onference on {M}ultimedia {I}nformation {R}etrieval 2008, {O}ctober 27-{N}ovember 01, 2008, {V}ancouver, {BC}, {C}anada}, address = {{V}ancouver, {CANADA}}, month = {10}, url = {http://www.eurecom.fr/publication/2570} }
See also: