26,717 Hits in 9.1 sec

Aggregation of Multiple Pseudo-Relevance Feedbacks for Image Search Reranking

Wei-Chao Lin
2019 IEEE Access  
Image retrieval effectiveness can be improved by pseudo relevance feedback (PRF), which automatically uses top-k images of the initial retrieval result as the pseudo feedback.  ...  INDEX TERMS Image retrieval, re-ranking, pseudo relevance feedback, Borda count.  ...  INTRODUCTION Content-based image retrieval (CBIR) focuses on automatically extracting low-level image features, such as color, texture, and shape, to index images.  ... 
doi:10.1109/access.2019.2942142 fatcat:jqucuyx4vrcqpg7jzv7vyv3ndu

A relevance feedback image retrieval scheme using multi-instance and pseudo-image concepts

Feng-Cheng Chang, Hsueh-Ming Hang, Rainer W. Lienhart, Noboru Babaguchi, Edward Y. Chang
2005 Storage and Retrieval Methods and Applications for Multimedia 2005  
In realizing the preceding concepts, an image search scheme is developed using the weighted low-level image features.  ...  The second key concept is that when the user does not provide a sufficient number of samples, how we generate a set of consistent "pseudo images".  ...  To simplify the simulation and clarify the effectiveness of the proposed estimation scheme, the goal of our system is to use only low-level features for automatically high-volume feature extraction and  ... 
doi:10.1117/12.586678 dblp:conf/spieSR/ChangH05 fatcat:fsu7jkgbo5dotmmnjd4m2shdju

A Relevance Feedback Image Retrieval Scheme Using Multi-Instance and Pseudo Image Concepts

2006 IEICE transactions on information and systems  
In realizing the preceding concepts, an image search scheme is developed using the weighted low-level image features.  ...  The second key concept is that when the user does not provide a sufficient number of samples, how we generate a set of consistent "pseudo images".  ...  To simplify the simulation and clarify the effectiveness of the proposed estimation scheme, the goal of our system is to use only low-level features for automatically high-volume feature extraction and  ... 
doi:10.1093/ietisy/e89-d.5.1720 fatcat:ehj4dykoefdgphnggfdzfxtubu

An Efficient Algorithm to Reduce the Semantic Gap between Image Contents and Tags

Grishma Y.Bobhate, Usha A. Jogalekar
2013 International Journal of Computer Applications  
Therefore, our presented research is going to reduce the problem of semantic gap by applying techniques to extract low level features of an image such as color, texture and edge.  ...  Then, construction of a mixed graph between image and tag to perform random walk on graph for getting accurate results in an efficient way. Experimental results show the effectiveness of our approach.  ...  It extracts low level features of images such as color, texture, shape to retrieve the images from a large collection of images.  ... 
doi:10.5120/12541-9136 fatcat:eakjpf5kt5h53eapjdls3rbpwu

Towards Data-Adaptive and User-Adaptive Image Retrieval by Peer Indexing

Jun Yang, Qing Li, Yueting Zhuang
2004 International Journal of Computer Vision  
Based on two-level image peer indices, retrieval parameters including query vectors and similarity metric can be optimized towards both data and user characteristics by applying the pseudo feedback strategy  ...  Extensive experiments have been conducted on real-world images to verify the effectiveness of our approach.  ...  The authors would like to express their thanks to Dr.Liu Wenyin for a fruitful discussion on the issue of user personalization, which helped our presentation on the user-adaptation aspect of this paper  ... 
doi:10.1023/b:visi.0000004836.59343.e9 fatcat:pfgkrtedhbfv3eeru2gkcgab6y

A Comparative Study of Dimension Reduction Techniques for Content-Based Image Retrivel

Sasikala G, Kowsalya R, Punithavalli M
2010 The International Journal of Multimedia & Its Applications  
Efficient and effective retrieval techniques of images are desired because of the explosive growth of digital images.  ...  Content-based image retrieval is a promising approach because of its automatic indexing and retrieval based on their semantic features and visual appearance.  ...  Relevance feedback is one of the most important techniques for narrowing down the gap between low-level visual features and high-level semantic concepts [3] .  ... 
doi:10.5121/ijma.2010.2303 fatcat:eebfcnngjfgerahgxq3zzuvyiq

Pseudo relevance feedback with incremental learning for high level feature detection

Shaoxi Xu, Sheng Tang, Jintao Li, Yongdong Zhang
2009 2009 IEEE International Conference on Multimedia and Expo  
Pseudo Relevance Feedback (PRF) has shown effective performance in information retrieval, but it has seldom been applied in the area of high level feature detection (HLF).  ...  We evaluate our approaches on the benchmark of TRECVID2008.  ...  The process of selecting pseudo positive samples in this method only utilizes the low level features for the distance comparison while ignores those semantic information embedding in positive training  ... 
doi:10.1109/icme.2009.5202566 dblp:conf/icmcs/XuTLZ09 fatcat:n4edntj5vraa3cvtel7itdzep4

Inter-media Pseudo-relevance Feedback Application to ImageCLEF 2006 Photo Retrieval [chapter]

Nicolas Maillot, Jean-Pierre Chevallet, Joo Hwee Lim
2007 Lecture Notes in Computer Science  
We show in particular how inter-media re-ranking and pseudo-relevance feedback have been used for producing the results. We have also tested Latent Semantic Analysis (LSA) approach on visual runs.  ...  This paper provides a description of the way results has been produced. The text/image database used is IAPR [1]. We have tested a cooperative use of a text retrieval and an image retrieval engine.  ...  In this case the image is split in 5x5 patches.The low-level features extracted on patches are the following: -Texture features used by our system are Gabor features [5] .  ... 
doi:10.1007/978-3-540-74999-8_92 fatcat:wu3wfpzjcjevpafind6lntsqvi

Content-Based Image Retrieval [chapter]

Borko Furht, Stephen W. Smoliar, HongJiang Zhang
1995 Video and Image Processing in Multimedia Systems  
The ability to search through images based on their content rather than on their low-level features is becoming more important as the number of available images grows.  ...  Index Terms-Content-based image retrieval, relevance feedback, region-based image retrieval, high-level semantics  ...  For example, the low level features in an image may represent that there is a brown object in the upperleft corner of an image, while the desired high level features would represent that there is a dog  ... 
doi:10.1007/978-1-4615-2277-5_11 fatcat:cdz4fglm4bdoxirkmxpaklixla

Generalized Manifold-Ranking-Based Image Retrieval

J. He, M. Li, H.-J. Zhang, H. Tong, C. Zhang
2006 IEEE Transactions on Image Processing  
Index Terms-Image retrieval, manifold ranking, outside the database, relevance feedback.  ...  Systematic experiments on a general-purpose image database consisting of 5 000 Corel images demonstrate the superiority of gMRBIR over state-of-the-art techniques.  ...  ACKNOWLEDGMENT The authors would like to thank S. Yan, X. Zheng, L. Zhang, and X.  ... 
doi:10.1109/tip.2006.877491 pmid:17022278 fatcat:xchcdifkkbemxb2xr3odubzmfi

Improving Retrieval Quality Using Pseudo Relevance Feedback in Content-Based Image Retrieval

Dinesha Chathurani Nanayakkara Wasam Uluwitige, Timothy Chappell, Shlomo Geva, Vinod Chandran
2016 Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16  
Pseudo-relevance feedback (PRF) has proven to be an effective mechanism for improving retrieval accuracy.  ...  Empirical evaluations based on standard benchmarks, namely Wang, Oliva & Torralba, and Corel datasets demonstrate the effectiveness of the proposed RB-PRF mechanism in image retrieval.  ...  The feedback vectors are then added together independently and two vectors are generated -one from the pseudo-relevant signatures and one from the pseudo-irrelevant signatures.  ... 
doi:10.1145/2911451.2914747 dblp:conf/sigir/UluwitigeCGC16 fatcat:ohil4r6xnvcinn5uqdjzdiw42i

A New Re-Ranking Method Using Enhanced Pseudo-Relevance Feedback for Content-Based Medical Image Retrieval

Yonggang HUANG, Jun ZHANG, Yongwang ZHAO, Dianfu MA
2012 IEICE transactions on information and systems  
We propose a novel re-ranking method for content-based medical image retrieval based on the idea of pseudo-relevance feedback (PRF).  ...  In step 2, after estimating a relevance probability for each of the highest ranked images, a fuzzy SVM ensemble based approach is adopted to re-rank the images.  ...  Acknowledgements The authors thank Dr. Ken Nakamura and the anonymous reviewer for valuable feedback to this work.  ... 
doi:10.1587/transinf.e95.d.694 fatcat:23qvun7ojzhr7k755htpvggsja

A Pseudo-Labeling Framework for Content-based Image Retrieval

Kim-Hui Yap, Kui Wu
2007 2007 IEEE Symposium on Computational Intelligence in Image and Signal Processing  
Experimental results based on a database of 10,000 images demonstrate the effectiveness of the proposed method.  ...  One of the main issues associated with relevance feedback in CBIR systems is the small sample problem where only a limited number of labeled samples are available for learning.  ...  These low-level features, however, may not correspond to the users' dynamic and subjective interpretation of image contents under various circumstances.  ... 
doi:10.1109/ciisp.2007.369179 fatcat:wi4kmdqwqzctdnsmt3nwde7w7a

Zero-Example Event Search using MultiModal Pseudo Relevance Feedback

Lu Jiang, Teruko Mitamura, Shoou-I Yu, Alexander G. Hauptmann
2014 Proceedings of International Conference on Multimedia Retrieval - ICMR '14  
The approach is unique in that it leverages not only semantic features, but also non-semantic low-level features for event search in the absence of training data.  ...  We propose a novel method MultiModal Pseudo Relevance Feedback (MMPRF) for event search in video, which requires no search examples from the user.  ...  DCNN features are 1000 visual objects trained on about 1.2 million ImageNet images by DCNN [12] . Two types of low-level features were used.  ... 
doi:10.1145/2578726.2578764 dblp:conf/mir/JiangMYH14 fatcat:4rrjs6y7cbfsfcjvjluvugv7sa

Visual Reranking: From Objectives to Strategies

xinmei tian, Dacheng Tao
2011 IEEE Multimedia  
The first disadvantage is the well-known semantic gap between low-level visual features and high-level semantic concepts, leading to irrelevant images returned.  ...  Third, textual information is insufficient to distinguish images of different relevance, which means that some slightly relevant samples will be returned as the results.  ...  Richter et al. proposed to model the similarity between images from multimodal cues, 14 specifically measuring similarity on the basis of low-level visual features and textual features (user tags).  ... 
doi:10.1109/mmul.2011.36 fatcat:ti34miqu4vbqnhto3oa6q7znpu
« Previous Showing results 1 — 15 out of 26,717 results