• Home
  • Researches
  • Local Detectors and Compact Descriptors for Visual Search

Local Detectors and Compact Descriptors for Visual Search

Local Visual detectors and descriptors have been studied for many years, but their applications, for example on mobile visual search, for large volume, low cost, low power embedded systems have been limited or negligible to date. One reason is the lack of an industry standard applicable at worldwide level. MPEG Compact Descriptors for Visual Search (CDVS) working group is filling this gap by defining a highly performing extraction stage and the bitstream syntax at its output in order to achieve interoperability between different implementation of clients and servers. As a contribution to such process, in our previous work we presented a review of existing methods for gray-levels interest point detection and description. This article extends our previous analysis in the framework of the MPEG CDVS Test Model considering further detectors and the use of color descriptors.

Dataset

Publications

1.

Local Detectors and Compact Descriptors for Visual Search: a Quantitative Comparison
(Simone Bianco, Davide Mazzini, Danilo Pietro Pau, Raimondo Schettini) In Digital Signal Processing, volume 44, pp. 1-13, Elsevier, 2015.

Download BibTex Doi Project Page 

@article{bianco2015local-detectors,
 author = {Bianco, Simone and Mazzini, Davide and Pau, Danilo Pietro and Schettini, Raimondo},
 year = {2015},
 pages = {1-13},
 title = {Local Detectors and Compact Descriptors for Visual Search: a Quantitative Comparison},
 volume = {44},
 publisher = {Elsevier},
 journal = {Digital Signal Processing},
 pdf = {/download/bianco2015local-detectors.pdf},
 doi = {10.1016/j.dsp.2015.06.001},
 projectref = {http://www.ivl.disco.unimib.it/activities/local-detectors/}}
2.

Quantitative Review of Local Descriptors for Visual Search
(Simone Bianco, Davide Mazzini, Danilo Pietro Pau, Raimondo Schettini) In IEEE International Conference on Consumer Electronics - Berlin (ICCE-Berlin), pp. 98-102, 2013.

BibTex Doi Project Page 

@inproceedings{bianco2013quantitative-review,
 author = {Bianco, Simone and Mazzini, Davide and Pau, Danilo Pietro and Schettini, Raimondo},
 year = {2013},
 pages = {98-102},
 title = {Quantitative Review of Local Descriptors for Visual Search},
 booktitle = {IEEE International Conference on Consumer Electronics - Berlin (ICCE-Berlin)},
 doi = {10.1109/ICCE-Berlin.2013.6697972},
 projectref = {http://www.ivl.disco.unimib.it/research/supermarketMilan/index.html}}
3.

Image retrieval system with disambiguation methods based on color descriptors
(Danilo Pietro Pau, Davide Mazzini, Raimondo Schettini, Simone Bianco) filed in 2013.

BibTex 

@patent{pau2013image-retrieval,
 author = {Pau, Danilo Pietro and Mazzini, Davide and Schettini, Raimondo and Bianco, Simone},
 title = {Image retrieval system with disambiguation methods based on color descriptors},
 yearfiled = {2013}}