259 Hits in 1.8 sec

Scalable triangulation-based logo recognition

Yannis Kalantidis, Lluis Garcia Pueyo, Michele Trevisiol, Roelof van Zwol, Yannis Avrithis
2011 Proceedings of the 1st ACM International Conference on Multimedia Retrieval - ICMR '11  
We propose a scalable logo recognition approach that extends the common bag-of-words model and incorporates local geometry in the indexing process.  ...  We evaluate our approach on a large-scale logo recognition dataset with more than four thousand classes.  ...  Finally, we intent to exploit the speed and scalability of the proposed approach for logo detection in videos.  ... 
doi:10.1145/1991996.1992016 dblp:conf/mir/KalantidisPTZA11 fatcat:fvghbgtzczaavasxefrbhna2qe

Multi-perspective cross-class domain adaptation for open logo detection

Hang Su, Shaogang Gong, Xiatian Zhu
2020 Computer Vision and Image Understanding  
This restricts their scalability to a large number of logo classes subject to limited labelling budget.  ...  In this work, we consider a more scalable open logo detection problem where only a fraction of logo classes are fully labelled whilst the remaining classes are only annotated with a clean icon image (e.g  ...  How to use the already labelled logo detection data in a scalable manner seems a promising approach.  ... 
doi:10.1016/j.cviu.2020.103156 fatcat:o42uzsff2rd4rkpjurvugcpn6a

Image retrieval based on spatial context with Relaxed Gabriel Graph pyramid

Xiaomeng Wu, Kunio Kashino
2014 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
Imposing the coherence of the spatial context on local features is becoming a necessity for object retrieval and recognition.  ...  Motivated by the success of proximity graphs in topological decomposition, clustering, and gradient estimation, we introduce a variation on and a generalization of Delaunay Triangulation, called a Relaxed  ...  INTRODUCTION The Bag-Of-Words (BOW) model based on local features has been shown to be successful in object retrieval and recognition along with its extensions [1, 2, 3, 4, 5, 6] .  ... 
doi:10.1109/icassp.2014.6854933 dblp:conf/icassp/WuK14a fatcat:d3hzlcp2nzeexelce5l3z7epku

Searching visual instances with topology checking and context modeling

Wei Zhang, Chong-Wah Ngo
2013 Proceedings of the 3rd ACM conference on International conference on multimedia retrieval - ICMR '13  
Specifically, we explore the use of (1) an elastic spatial topology checking technique based on Delaunay Triangulation (DT), and (2) a practical background context modeling method by simulating the "stare  ...  Based on the Bag-of-Words model, we propose two techniques tailored for Instance Search.  ...  (9053: Coca-cola logo, 9068: PUMA logo).  ... 
doi:10.1145/2461466.2461477 dblp:conf/mir/ZhangN13 fatcat:efo7g3f4lvhirdxcfmqojwgqxu

Compressive Change Retrieval for Moving Object Detection [article]

Tomoya Murase, Kanji Tanaka
2016 arXiv   pre-print
To address the above issues, we also present a practical change detection algorithm that uses compressed bag-of-words (BoW) image representation as a scalable solution.  ...  First, the retrieved reference images may frequently contain non-relevant reference images, because even state-of-the-art place-recognition techniques suffer from retrieval noise.  ...  To this end, we adapt triangulation-based analysis of LG, which was originally proposed for a different application, specifically, "scalable logo recognition", in [8] .  ... 
arXiv:1608.02051v1 fatcat:aznjsxur4jfwfb7dwt52os5usm

Bundle min-Hashing

Stefan Romberg, Rainer Lienhart
2013 International Journal of Multimedia Information Retrieval  
The recognition of logos in novel images is then performed by querying a database of reference images.  ...  We demonstrate the benefits of these techniques for both small object retrieval and logo recognition.  ...  Logo recognition Now that we have discussed visual features, vocabularies, feature bundling, re-ranking and synthetic query expansion; we present our final logo recognition system: Indexing The logo classes  ... 
doi:10.1007/s13735-013-0040-x fatcat:3jzzxkfvnjbizieidjusag2kyu


Wei Zhang, Lei Pang, Chong-Wah Ngo
2012 Proceedings of the 20th ACM international conference on Multimedia - MM '12  
., fashion, art, food, pet, logo, and landmark) over various QA categories (e.g., factoid, definition, how-to, and opinion).  ...  Figure 4 : 4 Construction of triangulation meshes based on the matching visual words between two images. The matched words are indicated with the same color.  ...  This result indicates the robustness and scalability of DT.  ... 
doi:10.1145/2393347.2393432 dblp:conf/mm/ZhangPN12 fatcat:ipaqc6w7dbeblezxe5gouyziey

Feature Fusion by Similarity Regression for Logo Retrieval

Fan Yang, Mayank Bansal
2015 2015 IEEE Winter Conference on Applications of Computer Vision  
We propose a simple yet effective multi-feature fusion approach based on regression models for logo retrieval.  ...  Logo class information from the training samples can also be included in the training process by learning an ensemble of regression models for individual logo classes.  ...  We experiment with two logo datasets, FlickrL-ogo32 [22] and BelgaLogo [11] . FlickrLogo32 contains 32 brand logo classes used for logo detection, recognition and retrieval tasks.  ... 
doi:10.1109/wacv.2015.132 dblp:conf/wacv/YangB15 fatcat:2o52zajdgbbbrasvejochlydzi

Establishing Multi-cast Groups in Computational Robotic Materials

Shang Ma, Homa Hosseinmardi, Nicholas Farrow, Richard Han, Nikolaus Correll
2012 2012 IEEE International Conference on Green Computing and Communications  
We describe our Bloom filter-based multicast communication (BMC) protocol, and report experimental results using a 48node Computational Robotic Material test-bed engaged in shape and gesture recognition  ...  In previous work, we proposed a Bloom filter-based approach to label the multicast group with an approximate error-resilient multicast tag that captures the temporal and spatial characteristics of the  ...  Taking advantage of the multicast group, e.g., for performing a consensus-based shape recognition, will therefore require some prior information on the size of the shape.  ... 
doi:10.1109/greencom.2012.74 dblp:conf/greencom/MaHFHC12 fatcat:im3locou4vgsdbrudwugvb33gu

PixSearcher: Searching Similar Images in Large Image Collections through Pixel Descriptors [chapter]

Tuan Nhon Dang, Leland Wilkinson
2014 Lecture Notes in Computer Science  
These descriptors are computed based on proximity graphs that are subsets of the Delaunay triangulation.  ...  These features are used for object recognition. The recognition accuracy can be improved by incorporating other image aspects such as color and texture.  ... 
doi:10.1007/978-3-319-14364-4_70 fatcat:mq7s2mvlg5fo5djkhjthuafpa4

Pharmapack: Mobile Fine-Grained Recognition Of Pharma Packages

Oscar Dabrowski, Taras Holotyak, Shideh Rezaeifar, Jonathan Schlechten, Olga Taran, Sviatoslav Voloshynovskiy
2018 Zenodo  
very rich and represents a mixture of text, logos and rarely some images.  ...  Additionally, this dataset corresponds to a typical production chain of consumer goods that should be well suited for future scalability in mobile recognition applications. B.  ... 
doi:10.5281/zenodo.1159779 fatcat:boc4egtouna3rhtbcrqmrodzba

Second-Order Configuration of Local Features for Geometrically Stable Image Matching and Retrieval

Xiaomeng Wu, Kunio Kashino
2015 IEEE transactions on circuits and systems for video technology (Print)  
We comprehensively evaluated our approach using Flickr Logos 32, Holiday, Oxford Buildings and Flickr 100K benchmarks.  ...  In this paper, we propose a Centrality-Sensitive Pyramid approach based on Delaunay triangulation for this purpose, which is described in Section V.  ...  [5] proposed the use of Delaunay triangulation (DT) for this purpose.  ... 
doi:10.1109/tcsvt.2014.2382985 fatcat:z473eu53pjbl5oxbkldjtb5ndq

Semantic analysis of soccer video using dynamic Bayesian network

Chung-Lin Huang, Huang-Chia Shih, Chung-Yuan Chao
2006 IEEE transactions on multimedia  
Different from previous shot-based semantic analysis approaches, the proposed semantic analysis is frame-based for each input frame, it provides the current semantics of the event nodes as well as the  ...  Based on BN/DBN, it can identify the special events in soccer games such as goal event, corner kick event, penalty kick event, and card event.  ...  A knowledge-based semantic inference scheme for events recognition in sports video has been presented by three-layer semantic inference scheme [19] .  ... 
doi:10.1109/tmm.2006.876289 fatcat:ghg7w6swyrgnrltqgdwe2lsgpa

Instance search retrospective with focus on TRECVID

George Awad, Wessel Kraaij, Paul Over, Shin'ichi Satoh
2017 International Journal of Multimedia Information Retrieval  
In the meantime commercial applications of these techniques such as logo recognition in sports TV coverage or the recognition of landmarks, wine labels, books by your mobile phone camera [7] had become  ...  face recognition [68] , and object detection [22] .  ... 
doi:10.1007/s13735-017-0121-3 pmid:28758054 pmcid:PMC5531298 fatcat:3khp2cscmbhohipfx246gspqlq

Development of an autonomous object transfer system by an unmanned aerial vehicle based on binocular vision

Xu Liu, Bo Chen, Yuqing He, Decai Li
2020 International Journal of Advanced Robotic Systems  
of the competition was to build a simulated tower using prefabricated components by an unmanned rotorcraft, which could be decomposed into the following four subtasks: (1) navigation and control, (2) recognition  ...  Recognition. The existing algorithm for target recognition can roughly divided into two categories: one is the traditional, 11-15 the other is deep learning based.  ...  Then the depth d of the target point can be figured out by the triangulation.  ... 
doi:10.1177/1729881420907732 fatcat:ji4u7tpsezdudferlxhtxkyi5q
« Previous Showing results 1 — 15 out of 259 results