The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Dalam makalah ini, kami mencadangkan algoritma carian visual mudah alih yang sangat cekap. Untuk proses pengekstrakan deskriptor, kami mencadangkan pengesanan ciri kerumitan rendah yang menggunakan titik utama tempatan yang dikesan bagi oktaf kasar untuk membimbing pembinaan ruang skala dan pengesanan ciri dalam oktaf halus. Operasi Gaussian dan Laplacian dilangkau untuk kawasan yang tidak penting, dan dengan itu masa pengkomputeran dijimatkan. Selain itu, pemilihan ciri diletakkan sebelum pengkomputeran orientasi untuk mengurangkan lagi kerumitan pengesanan ciri dengan pra-membuang beberapa titik tempatan yang tidak penting. Untuk proses mendapatkan semula imej, kami mereka bentuk kaedah penyusunan semula berprestasi tinggi, yang menggabungkan kedua-dua skor padanan deskriptor global dan skor persamaan deskriptor tempatan (LDSS). Dalam pengiraan LDSS, tf-idf pemadanan histogram berwajaran dilakukan untuk menyepadukan maklumat statistik pangkalan data. Keputusan menunjukkan bahawa pendekatan yang sangat cekap yang dicadangkan mencapai prestasi yang setanding dengan carian visual mudah alih yang terkini, manakala kerumitan pengekstrakan deskriptor dikurangkan sebahagian besarnya.
Chuang ZHU
Beijing University of Posts and Telecommunications
Xiao Feng HUANG
NVIDIA Corporation
Guo Qing XIANG
Peking University
Hui Hui DONG
Beijing University of Posts and Telecommunications
Jia Wen SONG
Peking University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Chuang ZHU, Xiao Feng HUANG, Guo Qing XIANG, Hui Hui DONG, Jia Wen SONG, "Highly Efficient Mobile Visual Search Algorithm" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 12, pp. 3073-3082, December 2018, doi: 10.1587/transinf.2018EDP7075.
Abstract: In this paper, we propose a highly efficient mobile visual search algorithm. For descriptor extraction process, we propose a low complexity feature detection which utilizes the detected local key points of the coarse octaves to guide the scale space construction and feature detection in the fine octave. The Gaussian and Laplacian operations are skipped for the unimportant area, and thus the computing time is saved. Besides, feature selection is placed before orientation computing to further reduce the complexity of feature detection by pre-discarding some unimportant local points. For the image retrieval process, we design a high-performance reranking method, which merges both the global descriptor matching score and the local descriptor similarity score (LDSS). In the calculating of LDSS, the tf-idf weighted histogram matching is performed to integrate the statistical information of the database. The results show that the proposed highly efficient approach achieves comparable performance with the state-of-the-art for mobile visual search, while the descriptor extraction complexity is largely reduced.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7075/_p
Salinan
@ARTICLE{e101-d_12_3073,
author={Chuang ZHU, Xiao Feng HUANG, Guo Qing XIANG, Hui Hui DONG, Jia Wen SONG, },
journal={IEICE TRANSACTIONS on Information},
title={Highly Efficient Mobile Visual Search Algorithm},
year={2018},
volume={E101-D},
number={12},
pages={3073-3082},
abstract={In this paper, we propose a highly efficient mobile visual search algorithm. For descriptor extraction process, we propose a low complexity feature detection which utilizes the detected local key points of the coarse octaves to guide the scale space construction and feature detection in the fine octave. The Gaussian and Laplacian operations are skipped for the unimportant area, and thus the computing time is saved. Besides, feature selection is placed before orientation computing to further reduce the complexity of feature detection by pre-discarding some unimportant local points. For the image retrieval process, we design a high-performance reranking method, which merges both the global descriptor matching score and the local descriptor similarity score (LDSS). In the calculating of LDSS, the tf-idf weighted histogram matching is performed to integrate the statistical information of the database. The results show that the proposed highly efficient approach achieves comparable performance with the state-of-the-art for mobile visual search, while the descriptor extraction complexity is largely reduced.},
keywords={},
doi={10.1587/transinf.2018EDP7075},
ISSN={1745-1361},
month={December},}
Salinan
TY - JOUR
TI - Highly Efficient Mobile Visual Search Algorithm
T2 - IEICE TRANSACTIONS on Information
SP - 3073
EP - 3082
AU - Chuang ZHU
AU - Xiao Feng HUANG
AU - Guo Qing XIANG
AU - Hui Hui DONG
AU - Jia Wen SONG
PY - 2018
DO - 10.1587/transinf.2018EDP7075
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2018
AB - In this paper, we propose a highly efficient mobile visual search algorithm. For descriptor extraction process, we propose a low complexity feature detection which utilizes the detected local key points of the coarse octaves to guide the scale space construction and feature detection in the fine octave. The Gaussian and Laplacian operations are skipped for the unimportant area, and thus the computing time is saved. Besides, feature selection is placed before orientation computing to further reduce the complexity of feature detection by pre-discarding some unimportant local points. For the image retrieval process, we design a high-performance reranking method, which merges both the global descriptor matching score and the local descriptor similarity score (LDSS). In the calculating of LDSS, the tf-idf weighted histogram matching is performed to integrate the statistical information of the database. The results show that the proposed highly efficient approach achieves comparable performance with the state-of-the-art for mobile visual search, while the descriptor extraction complexity is largely reduced.
ER -