Statistical Score Fusion for 3D Object Retrieval. Akguel, C. B., Sankur, B., & Yemez, Y. In 2008 IEEE 16TH SIGNAL PROCESSING, COMMUNICATION AND APPLICATIONS CONFERENCE, VOLS 1 AND 2, pages 284-287, 2008. IEEE. IEEE 16th Signal Processing and Communications Applications Conference, Aydin, TURKEY, APR 20-22, 2008
abstract   bibtex   
In this work, we introduce the score fusion problem for 3D object retrieval. Ongoing research in 3D object retrieval shows that no single descriptor is capable of providing fine grain discrimination required by prospective 3D search engines. We present a fusion algorithm that linearly combines similarity information originating from multiple shape descriptors. We learn the optimal set of weights in the linear combination by minimizing the emprical ranking risk. The algorithm is based on a recently introduced rigorous statistical ranking framework, for which consistency and fast rate of convergence of empirical ranking risk minimizers have been established. We report the results of relevance feedback search on a large 3D object database, the Princeton Shape Benchmark. Experiments show that, under query formulations with user intervention, the proposed score fusion scheme boosts the performance of the 3D retrieval machine significantly.
@inproceedings{ ISI:000261359200071,
Author = {Akguel, Ceyhun Burak and Sankur, Buelent and Yemez, Yuecel},
Book-Group-Author = {{IEEE}},
Title = {{Statistical Score Fusion for 3D Object Retrieval}},
Booktitle = {{2008 IEEE 16TH SIGNAL PROCESSING, COMMUNICATION AND APPLICATIONS
   CONFERENCE, VOLS 1 AND 2}},
Year = {{2008}},
Pages = {{284-287}},
Note = {{IEEE 16th Signal Processing and Communications Applications Conference,
   Aydin, TURKEY, APR 20-22, 2008}},
Organization = {{IEEE}},
Abstract = {{In this work, we introduce the score fusion problem for 3D object
   retrieval. Ongoing research in 3D object retrieval shows that no single
   descriptor is capable of providing fine grain discrimination required by
   prospective 3D search engines. We present a fusion algorithm that
   linearly combines similarity information originating from multiple shape
   descriptors. We learn the optimal set of weights in the linear
   combination by minimizing the emprical ranking risk. The algorithm is
   based on a recently introduced rigorous statistical ranking framework,
   for which consistency and fast rate of convergence of empirical ranking
   risk minimizers have been established. We report the results of
   relevance feedback search on a large 3D object database, the Princeton
   Shape Benchmark. Experiments show that, under query formulations with
   user intervention, the proposed score fusion scheme boosts the
   performance of the 3D retrieval machine significantly.}},
ISBN = {{978-1-4244-1998-2}},
Unique-ID = {{ISI:000261359200071}},
}

Downloads: 0