Filters








179,061 Hits in 3.0 sec

Diverse Image Annotation

Baoyuan Wu, Fan Jia, Wei Liu, Bernard Ghanem
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
In this work, we study a new image annotation task called diverse image annotation (DIA).  ...  As compared to the conventional image annotation task, DIA requires the tags to be not only representative of the image but also diverse from each other, so as to reduce redundancy.  ...  (i) We propose a new task called diverse image annotation.  ... 
doi:10.1109/cvpr.2017.656 dblp:conf/cvpr/WuJLG17 fatcat:semd4duujrgcvg27dhpi2fvwsy

Tagging Like Humans: Diverse and Distinct Image Annotation

Baoyuan Wu, Weidong Chen, Peng Sun, Wei Liu, Bernard Ghanem, Siwei Lyu
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
In this work we propose a new automatic image annotation model, dubbed diverse and distinct image annotation (D 2 IA).  ...  Multiple such tag subsets that cover diverse semantic aspects or diverse semantic levels of the image contents are generated by randomly perturbing the DPP sampling process.  ...  In this work, we propose a novel image annotation model, namely diverse and distinct image annotation (D 2 IA), which aims to improve the diversity and distinctiveness of the tags for an image by learning  ... 
doi:10.1109/cvpr.2018.00831 dblp:conf/cvpr/WuCS0GL18 fatcat:seizrdfwqfcz7lr2ftcdilp2ri

Tagging like Humans: Diverse and Distinct Image Annotation [article]

Baoyuan Wu, Weidong Chen, Peng Sun, Wei Liu, Bernard Ghanem, Siwei Lyu
2018 arXiv   pre-print
In this work we propose a new automatic image annotation model, dubbed diverse and distinct image annotation (D2IA).  ...  Multiple such tag subsets that cover diverse semantic aspects or diverse semantic levels of the image contents are generated by randomly perturbing the DPP sampling process.  ...  In this work, we propose a novel image annotation model, namely diverse and distinct image annotation (D 2 IA), which aims to improve the diversity and distinctiveness of the tags for an image by learning  ... 
arXiv:1804.00113v1 fatcat:zn2jiuawhrhkbltnghr4bpmraq

Holistic and Comprehensive Annotation of Clinically Significant Findings on Diverse CT Images: Learning from Radiology Reports and Label Ontology [article]

Ke Yan, Yifan Peng, Veit Sandfort, Mohammadhadi Bagheri, Zhiyong Lu, Ronald M. Summers
2019 arXiv   pre-print
We evaluated LesaNet on the public DeepLesion dataset, which contains over 32K diverse lesion images.  ...  With the images and text-mined labels, we propose a lesion annotation network (LesaNet) based on a multilabel convolutional neural network (CNN) to learn all labels holistically.  ...  To learn to annotate lesions, a large-scale and diverse dataset of lesion images is needed. Existing lesion datasets [8, 34] are typically either too small or less diverse.  ... 
arXiv:1904.04661v2 fatcat:5e7eeoimujdetnpc2xzfo5kbjy

Image annotation by hierarchical mapping of features

Qiankun Zhao, Prasenjit Mitra, C. Lee Giles
2007 Proceedings of the 16th international conference on World Wide Web - WWW '07  
Moreover, we propose a novel annotation strategy that maximizes both the accuracy and the diversity of the generated annotation by generalizing or specifying the annotation in the corresponding annotation  ...  Experiments with 4500 scientific images from Royal Society of Chemistry journals show that the proposed annotation approach produces satisfactory results at different levels of annotations.  ...  One goal of the image annotation is to produce as diverse as possible annotations, We define the diversity of annotation as: A d = M ax( a i =a j ∈A Dist(ai, aj)) where A d is the diversity of the annotation  ... 
doi:10.1145/1242572.1242784 dblp:conf/www/ZhaoMG07 fatcat:n437s3pfxjforelxg534a6fhim

Adding semantics to the reliable object annotated image databases

Irfanullah, Nida Aslam, Jonathan Loo, Roohullah, Martin Loomes
2011 Procedia Computer Science  
Manual annotation is the only source to overwhelming this, which is not only time consuming and costly but also lacks semantic enrichment in terms of concept diversity and concept enrichability.  ...  Results demonstrate the substantial improvement in terms of concept diversity, concept enrichment and retrieval performance.  ...  Concept Diversity The Concept Diversity of annotations expresses the different topics or objects name exist in the dataset.  ... 
doi:10.1016/j.procs.2010.12.069 fatcat:fpyzantyyzdvberfe6k3xb5g7u

Predicting Foreground Object Ambiguity and Efficiently Crowdsourcing the Segmentation(s) [article]

Danna Gurari and Kun He and Bo Xiong and Jianming Zhang and Mehrnoosh Sameki and Suyog Dutt Jain and Stan Sclaroff and Margrit Betke and Kristen Grauman
2017 arXiv   pre-print
Specifically, we distinguish between images which lead multiple annotators to segment different foreground objects (ambiguous) versus minor inter-annotator differences of the same object.  ...  Experiments show our system eliminates up to 47% of human effort compared to existing crowdsourcing methods with no loss in capturing the diversity of ground truths.  ...  the first annotation for the k-th image (defined below), d ja represents the diversity captured by the a-th redundant annotation for the j-th image, and D(I) reflects the total annotation diversity captured  ... 
arXiv:1705.00366v1 fatcat:slubd4d3hnewtbl4leiklktgve

Annotation Efficient Person Re-Identification with Diverse Cluster-Based Pair Selection [article]

Lantian Xue, Yixiong Zou, Peixi Peng, Yonghong Tian, Tiejun Huang
2022 arXiv   pre-print
To solve this problem, we propose the Annotation Efficient Person Re-Identification method to select image pairs from an alternative pair set according to the fallibility and diversity of pairs, and train  ...  Specifically, we design an annotation and training framework to firstly reduce the size of the alternative pair set by clustering all images considering the locality of features, secondly select images  ...  Works [5, 4] inspired by active learning select few image pairs for human to annotate, but the annotation cost may still be related to the number of images.  ... 
arXiv:2203.05395v2 fatcat:fyivygvjazgwdpmwn7vcr4cv3y

Retrieving Diverse Social Images at MediaEval 2016: Challenge, Dataset and Evaluation

Bogdan Ionescu, Alexandru-Lucian Gînsca, Maia Zaharieva, Bogdan Boteanu, Mihai Lupu, Henning Müller
2016 MediaEval Benchmarking Initiative for Multimedia Evaluation  
This paper provides an overview of the Retrieving Diverse Social Images task that is organized as part of the Media-Eval 2016 Benchmarking Initiative for Multimedia Evaluation.  ...  The task addresses the problem of result diversification in the context of social photo retrieval where images, metadata, text information, user tagging profiles and content and text models are available  ...  GROUND TRUTH Both relevance and diversity annotations were carried out by expert annotators.  ... 
dblp:conf/mediaeval/IonescuGZBLM16 fatcat:ljsbzwt6orba3l46qkvvzowtju

Overview of the ImageCLEFphoto 2008 Photographic Retrieval Task [chapter]

Thomas Arni, Paul Clough, Mark Sanderson, Michael Grubinger
2009 Lecture Notes in Computer Science  
In 2008, the evaluation task concentrated on promoting diversity within the top 20 results from a multilingual image collection.  ...  Some of the findings include that the choice of annotation language is almost negligible and the best runs are by combining concept and content-based retrieval methods.  ...  . • Annotation completeness: each image caption exhibited the same level of annotation completenessthere were no images without annotations (as experimented with in 2006).  ... 
doi:10.1007/978-3-642-04447-2_62 fatcat:l552y7wq3bhcvlkzr5wfb3te54

Retrieving Diverse Social Images at MediaEval 2015: Challenge, Dataset and Evaluation

Bogdan Ionescu, Alexandru-Lucian Gînsca, Bogdan Boteanu, Adrian Popescu, Mihai Lupu, Henning Müller
2015 MediaEval Benchmarking Initiative for Multimedia Evaluation  
This paper provides an overview of the Retrieving Diverse Social Images task that is organized as part of the Media-Eval 2015 Benchmarking Initiative for Multimedia Evaluation.  ...  The task addresses the problem of result diversification and user annotation credibility estimation in the context of social photo retrieval.  ...  GROUND TRUTH Both relevance and diversity annotations were carried out by expert annotators with advanced knowledge of the loca-tion characteristics (mainly learned from last years' tasks and Internet  ... 
dblp:conf/mediaeval/IonescuGBPLM15 fatcat:utluxftg4rdelbwiwoklerbcdi

Retrieving Diverse Social Images at MediaEval 2014: Challenge, Dataset and Evaluation

Bogdan Ionescu, Adrian Popescu, Mihai Lupu, Alexandru-Lucian Gînsca, Henning Müller
2014 MediaEval Benchmarking Initiative for Multimedia Evaluation  
This paper provides an overview of the Retrieving Diverse Social Images task that is organized as part of the Media-Eval 2014 Benchmarking Initiative for Multimedia Evaluation.  ...  CONCLUSIONS The Retrieving Diverse Social Images task provides participants with a comparative and collaborative evaluation framework for social image retrieval techniques with explicit focus on result  ...  INTRODUCTION An efficient image retrieval system should be able to present results that are both relevant and that are covering diverse aspects of a query (e.g., sub-topics).  ... 
dblp:conf/mediaeval/IonescuPLGM14 fatcat:pxl2s2yzizdfpe3uwahowysgci

Div150Cred

Bogdan Ionescu, Adrian Popescu, Mihai Lupu, Alexandru Lucian Gînscă, Bogdan Boteanu, Henning Müller
2015 Proceedings of the 6th ACM Multimedia Systems Conference on - MMSys '15  
The dataset comes with associated relevance and diversity assessments performed by human annotators.  ...  The proposed dataset was validated during the 2014 Retrieving Diverse Social Images Task at the MediaEval Benchmarking Initiative.  ...  DATASET ANNOTATION Images are annotated for their relevance and diversity.  ... 
doi:10.1145/2713168.2713192 dblp:conf/mmsys/IonescuPLGBM15 fatcat:scpqmaxhgjh75n6o4em6o6is5a

Beyond Comparing Image Pairs: Setwise Active Learning for Relative Attributes

Lucy Liang, Kristen Grauman
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
It is useful to automatically compare images based on their visual properties-to predict which image is brighter, more feminine, more blurry, etc.  ...  On three challenging datasets and experiments with "live" online annotators, the proposed method outperforms both traditional passive learning as well as existing active rank learning methods.  ...  To learn relative attribute ranking functions, we propose an efficient active selection criterion that asks annotators to partially order a set of diverse yet informative images.  ... 
doi:10.1109/cvpr.2014.34 dblp:conf/cvpr/LiangG14 fatcat:fhy3nb7xjbhi7jhlmffxcbnl6e

Retrieving Diverse Social Images at MediaEval 2013: Objectives, Dataset and Evaluation

Bogdan Ionescu, María Menéndez, Henning Müller, Adrian Popescu
2013 MediaEval Benchmarking Initiative for Multimedia Evaluation  
This paper provides an overview of the Retrieving Diverse Social Images task that is organized as part of the Medi-aEval 2013 Benchmarking Initiative for Multimedia Evaluation.  ...  The testset was annotated by 7 persons, each expert annotated a different part of the data set leading in the end to 3 annotations per image.  ...  CONCLUSIONS The Retrieving Diverse Social Images Task provides participants with a comparative and collaborative evaluation framework for social image retrieval techniques with explicit focus on result  ... 
dblp:conf/mediaeval/IonescuMMP13 fatcat:pqn6s2zxpzgctf4ccg3i6v3bky
« Previous Showing results 1 — 15 out of 179,061 results