72,416 Hits in 5.6 sec

Relevance aggregation projections for image retrieval

Wei Liu, Wei Jiang, Shih-Fu Chang
2008 Proceedings of the 2008 international conference on Content-based image and video retrieval - CIVR '08  
In this paper, we address the two issues and propose a novel effective method called Relevance Aggregation Projections (RAP) for learning potent subspace projections in a semi-supervised way.  ...  To narrow the semantic gap in content-based image retrieval (CBIR), relevance feedback is utilized to explore knowledge about the user's intention in finding a target image or a image category.  ...  CONCLUSIONS In this paper, a new subspace learning technique, Relevance Aggregation Projections (RAP), for content-based image retrieval is proposed.  ... 
doi:10.1145/1386352.1386372 dblp:conf/civr/LiuJC08 fatcat:fdsfrs4rcnfvpfmnyuiqxagzke

X-Pool: Cross-Modal Language-Video Attention for Text-Video Retrieval [article]

Satya Krishna Gorti, Noel Vouitsis, Junwei Ma, Keyvan Golestan, Maksims Volkovs, Animesh Garg, Guangwei Yu
2022 arXiv   pre-print
Therefore, for a given text, a retrieval model should focus on the text's most semantically similar video sub-regions to make a more relevant comparison.  ...  In text-video retrieval, the objective is to learn a cross-modal similarity function between a text and a video that ranks relevant text-video pairs higher than irrelevant pairs.  ...  Our goal is to bootstrap from a pre-trained joint text-image model and extend it towards a joint text-video model for the task of text-video retrieval. Text-Video Retrieval.  ... 
arXiv:2203.15086v1 fatcat:cde3cco37jhhrpojotudm33ihu

Crossing textual and visual content in different application scenarios

Julien Ah-Pine, Marco Bressan, Stephane Clinchant, Gabriela Csurka, Yves Hoppenot, Jean-Michel Renders
2008 Multimedia tools and applications  
Our first method proposes using a mixture model of the aggregate components, considering them as a single relevance concept.  ...  We also introduce the monomodal similarity measures for text and images that serve as basic components for both proposed trans-media similarities.  ...  Acknowledgements The authors want to thank particularly INA for their contributions in our work and Florent Perronin for his greatly appreciated help in applying some of the Generic Visual Categorizer  ... 
doi:10.1007/s11042-008-0246-8 fatcat:w3ilxvdl65bgpl453z7izlnzbq

Visual Recognition in the EAGLE Project

Giuseppe Amato, Paolo Bolettieri, Fabrizio Falchi, Fausto Rabitti, Lucia Vadicamo
2015 Italian Information Retrieval Workshop  
In this paper, we present a system for visually retrieving ancient inscriptions, developed in the context of the ongoing Europeana network of Ancient Greek and Latin Epigraphy (EAGLE) EU Project.  ...  The experimental results show that the Vector of Locally Aggregated Descriptors is a promising encoding strategy for performing visual recognition in this specific context.  ...  Acknoledgments This work was partially supported by EAGLE (Europeana network of Ancient Greek and Latin Epigraphy, co-founded by the European Commision, CIP-ICT-PSP.2012.2.1 -Europeana and creativity, Project  ... 
dblp:conf/iir/AmatoBFRV15 fatcat:t45yxx5rmvefffiulacuitinae

Some Results Using Different Approaches to Merge Visual and Text-Based Features in CLEF'08 Photo Collection [chapter]

Ana García-Serrano, Xaro Benavent, Ruben Granados, José Miguel Goñi-Menoyo
2009 Lecture Notes in Computer Science  
Our main aim was to experiment with several merging approaches to fuse text-based retrieval and content-based retrieval results, and it happened that we improve the text-based baseline when applying one  ...  This paper describes the participation of the MIRACLE team 1 at the ImageCLEF Photographic Retrieval task of CLEF 2008. We succeeded in submitting 41 runs.  ...  Experiments applying ENRICH improved the baseline in MAP and in number of relevant images retrieved.  ... 
doi:10.1007/978-3-642-04447-2_69 fatcat:fr3yntnoyfcxvnsfhsxybpjave

ViSTA: Vision and Scene Text Aggregation for Cross-Modal Retrieval [article]

Mengjun Cheng, Yipeng Sun, Longchao Wang, Xiongwei Zhu, Kun Yao, Jie Chen, Guoli Song, Junyu Han, Jingtuo Liu, Errui Ding, Jingdong Wang
2022 arXiv   pre-print
Specifically, ViSTA utilizes transformer blocks to directly encode image patches and fuse scene text embedding to learn an aggregated visual representation for cross-modal retrieval.  ...  Visual appearance is considered to be the most important cue to understand images for cross-modal retrieval, while sometimes the scene text appearing in images can provide valuable information to understand  ...  Cross-modal text-to-image retrieval [10, 11, 22] aims to return the most relevant candidate based on the relevance between the text content of a query and the visual appearance of an image.  ... 
arXiv:2203.16778v1 fatcat:hldin76ql5hqtiq7ppvwmf6pmy

Immediate ROI Search for 3-D Medical Images [chapter]

Karen Simonyan, Marc Modat, Sebastien Ourselin, David Cash, Antonio Criminisi, Andrew Zisserman
2013 Lecture Notes in Computer Science  
image registration schemes; (iii) we propose a discriminative method for learning to rank the returned images based on the content of the ROI.  ...  The objective of this work is a scalable, real-time, visual search engine for 3-D medical images, where a user is able to select a query Region Of Interest (ROI) and automatically detect the corresponding  ...  Data collection and sharing for this project was funded by the Alzheimer's Disease Neuroimaging Initiative (ADNI) (NIH Grant U01 AG024904).  ... 
doi:10.1007/978-3-642-36678-9_6 fatcat:xy4bwwaybbgh3kdkqtvpayhgpy

Online social image ranking in diversified preferences

Xuezhuan Zhao, Lishen Pei, Tao Li, Zheng Zhang
2020 EURASIP Journal on Image and Video Processing  
AbstractDue to the prevalence of social media service, effective and efficient online image retrieval is in urgent need to satisfy diversified requirements of Web users.  ...  Last, we propose an effective and efficient position-sensitive rank aggregation approach to aggregate multiple ranking results based on the user preference specification.  ...  Acknowledgements Thanks to all those who have suggested and given guidance for this article.  ... 
doi:10.1186/s13640-020-00540-4 fatcat:mkva766gh5cdxbztsbcfiz6lse

Contextual Similarity Aggregation with Self-attention for Visual Re-ranking [article]

Jianbo Ouyang, Hui Wu, Min Wang, Wengang Zhou, Houqiang Li
2021 arXiv   pre-print
In image retrieval, it is observed that the contextual similarity among the top-ranked images is an important clue to distinguish the semantic relevance.  ...  In our approach, for each image in the top-K ranking list, we represent it into an affinity feature vector by comparing it with a set of anchor images.  ...  Specifically, in MHA, different projection matrices for Q, K, and V are used for different heads, and these matrices can project affinity features into different subspaces.  ... 
arXiv:2110.13430v1 fatcat:v2wc3mdbgjhuti32nqc25ruk5i

Leveraging Implicit Spatial Information in Global Features for Image Retrieval [article]

Pierre Jacob, David Picard, Aymeric Histace, Edouard Klein
2018 arXiv   pre-print
Most image retrieval methods use global features that aggregate local distinctive patterns into a single representation.  ...  However, the aggregation process destroys the relative spatial information by considering orderless sets of local descriptors.  ...  Conclusion In this paper, we propose the Improved Spatial Tensor Aggregation (ISTA) for aggregating local features into a single representation taylored for image retrieval.  ... 
arXiv:1806.08991v1 fatcat:pqs2ymlhavgdhg4zmxwtb2jfri

Similarity Adaptation in an Exploratory Retrieval Scenario [chapter]

Sebastian Stober, Andreas Nürnberger
2011 Lecture Notes in Computer Science  
Exploratory retrieval tools support users in search scenarios where the retrieval goal cannot be stated explicitly as a query or user rather want to browse a collection in order to get an overview and  ...  It uses a complex multi-focus fish-eye distortion of a projection to visualize neighborhood that is automatically adapted to the user's current focus of interest.  ...  W.r.t. the user's retrieval goal -finding five images for each topic -two performance value are of interest: the precisions at rank 5 and the number of new relevant images in secondary focus.  ... 
doi:10.1007/978-3-642-27169-4_11 fatcat:ujnzbltm2naojb6jaz6zlrgm6e

Detect-to-Retrieve: Efficient Regional Aggregation for Image Search [article]

Marvin Teichmann, Andre Araujo, Menglong Zhu, Jack Sim
2019 arXiv   pre-print
However, due to the lack of bounding-box datasets for objects of interest among retrieval benchmarks, most recent work on regional representations has focused on either uniform or class-agnostic region  ...  In addition, we introduce a novel regional aggregated selective match kernel (R-ASMK) to effectively combine information from detected regions into an improved holistic image representation.  ...  Conclusions In this paper, we present an efficient regional aggregation method for image retrieval.  ... 
arXiv:1812.01584v2 fatcat:dicbws7pdfaoxbd4pezucmmv7q

MIRACLE-FI at ImageCLEFphoto 2008: Experiences in merging Text-based and Content-based Retrievals

Ruben Granados, Xaro Benavent, Ana García-Serrano, José M. Goñi
2008 Conference and Labs of the Evaluation Forum  
In this is new participation of the group, our first purpose is to evaluate our own tools for text-based retrieval and for content-based retrieval using different similarity metrics and the aggregation  ...  The last one was used to select the relevant images to the content-based module. No clustering strategies were analyzed.  ...  Acknowledgements This work has been partially supported by the Spanish R+D National Plan, by means of the project BRAVO (Multilingual and Multimodal Answers Advanced Search -Information Retrieval), TIN2007  ... 
dblp:conf/clef/GranadosBGG08 fatcat:wcvoamcpz5durkso74u6chbsqy

IRAbMC: Image Recommendation with Absorbing Markov Chain

Sejal D, Rashmi V, Dinesh Anvekar, Venugopal K R, S S Iyengar, L M Patnaik
2015 2015 Annual IEEE India Conference (INDICON)  
In this paper, we present an algorithm Image Recommendation with Absorbing Markov Chain (JRAbMC) to retrieve relevant images for user input query.  ...  Image Recommendation is an important feature for search engine as tremendous amount images are available online. It is necessary to retrieve relevant images to meet user's requirement.  ...  ., [20] have proposed a semi supervised m � thod . call�d Maximum Margin Projection(MMP) for dImenSIOnalIty reduction, which focuses on local discriminant analysis for image retrieval.  ... 
doi:10.1109/indicon.2015.7443286 fatcat:kt55u3hbf5halbfdhude62bmq4

Boosting image retrieval through aggregating search results based on visual annotations

Ximena Olivares, Massimiliano Ciaramita, Roelof van Zwol
2008 Proceeding of the 16th ACM international conference on Multimedia - MM '08  
For this purpose we adopt the bag-of-visual-words approach for contentbased image retrieval as our baseline.  ...  image retrieval techniques to complement the text-based search.  ...  System S1 S2 S3 S4 S5 Number of Topics 30 30 30 30 30 Images Retrieved 750 750 750 742 748 Relevant 2187 2187 2187 2187 2187 Relevant Retrieved 393 149 301 494 562 MAP 0.12  ... 
doi:10.1145/1459359.1459386 dblp:conf/mm/OlivaresCZ08 fatcat:5ehhsdbdlbfwvaq4bn6z3ruzqm
« Previous Showing results 1 — 15 out of 72,416 results