Mobile Visual Search: Architectures, Technologies, and the Emerging MPEG Standard

Girod, Bernd; Chandrasekhar, Vijay; Grzeszczuk, Radek; Reznik, Yuriy A.

doi:10.1109/mmul.2011.48

Cited by 53 publications

(38 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fig. 6 (red o' s) shows search-memory complexity exponents of the binary system with the rate R = 0.5 and q = 0.1, using (22) and (25). The point (0.5, 0.5) is related to the exhaustive search scheme that we have R e = S e = R. As it was expected from Proposition 3, Fig.…”

Section: A Binary Sourcesupporting

confidence: 55%

“…The Markov chain U ↔ X ↔ Y is also utilized in an information retrieval setting introduced in [25], [26], where, to reduce the search complexity, the authors proposed clustering based on k-means over the database entries. For a given query, to find the most related contents they investigated clusters whose centroids are among the closest ones to the query.…”

Section: Non-excess Ratementioning

confidence: 99%

“…In Fig. 12 red o's show the achievable search-memory complexity exponent pairs for the Gaussian identification system with rate R = 0.5, using (25) and (22). Also in Fig.…”

Section: ) First Decoder As a Single-output Decodermentioning

confidence: 99%

See 2 more Smart Citations

Identification Rate, Search and Memory Complexity Tradeoff: Fundamental Limits

Farhadzadeh

Willems

2016

IEEE Trans. Inform. Theory

View full text Add to dashboard Cite

Abstract-In an information-theoretic framework, we introduce a two-stage decoding scheme capable of achieving identification capacity to address search and memory complexities in large-scale identification systems. This twostage decoding procedure is accomplished as follows. For a given query, at the first stage, a list of cluster indices is estimated. Then, at the second stage, refinement checks are performed for all members of the clusters to produce a single index. The first result this paper presents is the achievable rate quadruple region that specifies necessary conditions that the two-stage decoding scheme should satisfy to be able to achieve the identification capacity. The rest of paper is designated to investigate various achievable rate quadruples in which the proposed twostage identification setup can reduce the search complexity with respect to conventional identification setups.

show abstract

Section: A Binary Sourcesupporting

confidence: 55%

Section: Non-excess Ratementioning

confidence: 99%

See 1 more Smart Citation

Identification Rate, Search and Memory Complexity Tradeoff: Fundamental Limits

Farhadzadeh

Willems

2016

IEEE Trans. Inform. Theory

View full text Add to dashboard Cite

show abstract

“…The corresponding information about the object is sent out via wireless network. So this retrieval method is based on the mobile terminal and wireless network [3].…”

Section: The Mobile Visual Search Systemmentioning

confidence: 99%

The Mobile Visual Search Guiding System Based on SIFT

Zhou³

et al. 2016

IJFGCN

View full text Add to dashboard Cite

show abstract

“…Girod et al [2][3] proposed a visual search system in 2011. They developed new features and data structures for better feature extraction and retrieval.…”

Section: Related Workmentioning

confidence: 99%

Snap n' shop: Visual search-based mobile shopping made a breeze by machine and crowd intelligence

You

Yuan

Wang

et al. 2015

Proceedings of the 2015 IEEE 9th International Conference on Semantic Computing (IEEE ICSC 2015)

View full text Add to dashboard Cite

Abstract-The increasing popularity of smartphones has significantly changed the way we live. Today's powerful mobile systems provide us with all kinds of convenient services. Thanks to the wide variety of available apps, it has never been so easy for people to shop, to navigate, and to communicate with others. However, for some tasks we can further improve the user experience by employing newly developed algorithms. In this work, we try to improve visual search based mobile shopping experience by using machine and crowd intelligence. In particular, our system enables precise object selection, which would lead to more accurate visual search results. We also use crowdsourcing to further extend the system's prowess. We conduct experiments on user interface design and retrieval performance, which validate the effectiveness and ease of use of the proposed system. Meanwhile, components in the system are quite modular, allowing the flexibility of adding or improving different modules of the whole system.

show abstract

Mobile Visual Search: Architectures, Technologies, and the Emerging MPEG Standard

Cited by 53 publications

References 14 publications

Identification Rate, Search and Memory Complexity Tradeoff: Fundamental Limits

Identification Rate, Search and Memory Complexity Tradeoff: Fundamental Limits

The Mobile Visual Search Guiding System Based on SIFT

Snap n' shop: Visual search-based mobile shopping made a breeze by machine and crowd intelligence

Contact Info

Product

Resources

About