276 Hits in 5.7 sec

Incorporation Of Semantic Segmentation Information In Deep Hashing Techniques For Image Retrieval

Gkountakos Konstantinos, Semertzidis Theodoros, Papadopoulos Georgios Th., Daras Petros
2017 Zenodo  
In this paper, a novel approach to deep hashing is proposed, which incorporates local-level information, in the form of image semantic segmentation masks, during the hash code learning step.  ...  The proposed framework makes use of pixel-level classification labels, i.e. following a point-wise supervised learning methodology.  ...  The unsupervised approaches aim at learning feature representations based only on the statistics of the target data e.g. the variance of each dimension or its cardinality [8] .  ... 
doi:10.5281/zenodo.1076432 fatcat:uwqt7o4udbh75lygtlfevt7f6a

Randomized Locality Sensitive Vocabularies for Bag-of-Features Model [chapter]

Yadong Mu, Ju Sun, Tony X. Han, Loong-Fah Cheong, Shuicheng Yan
2010 Lecture Notes in Computer Science  
Integrating ideas from the Locality Sensitive Hashing (LSH) and the Random Forest (RF), RLSV generates and aggregates multiple visual vocabularies based on random projections, without taking clustering  ...  When visual data scale up (in terms of the dimensionality of features or/and the number of samples), most existing algorithms (e.g. k-means) become unfavorable due to the prohibitive time and space requirements  ...  For example, the popular SIFT descriptor [1] used for local visual feature description has 128 dimensions, while a typical image at normal resolution can produce 1k ∼ 2k such primitive feature descriptors  ... 
doi:10.1007/978-3-642-15558-1_54 fatcat:6ynd25pzmfapjhi7kl4v72f53i

Deep Metric and Hash-Code Learning for Content-Based Retrieval of Remote Sensing Images

Subhankar Roy, Enver Sangineto, Begum Demir, Nicu Sebe
2018 IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium  
Experiments on two aerial benchmark archives show that the proposed method outperforms previous state-of-theart hashing approaches by up to 5.4% using the same number of hash bits per image.  ...  other hand, encourage the network to produce final activation values (i.e., descriptors) that can be easily binarized.  ...  images based on the extracted descriptors.  ... 
doi:10.1109/igarss.2018.8518381 dblp:conf/igarss/RoySDS18 fatcat:bmq4flrinbbq7hbtgxewtejosq

Efficiently searching for similar images

Kristen Grauman
2010 Communications of the ACM  
Many problems demand fast, accurate search of very large databases of images, but often the most effective metrics for image comparisons do not mesh well with known efficient search methods.  ...  This is an extended version of an invited article that will appear in the Communications of the ACM (CACM) in 2009.  ...  We can think of each xi as a descriptor for one of the elliptical image regions on the koalas in Figure 2 .  ... 
doi:10.1145/1743546.1743570 fatcat:j4xgupwe6zgzbmzivpvkuyk6gy

Thick boundaries in binary space and their influence on nearest-neighbor search

Tomasz Trzcinski, Vincent Lepetit, Pascal Fua
2012 Pattern Recognition Letters  
A different approach to speeding up nearest-neighbor search is to binarize the real-valued descriptors using techniques such as Boosting [11], hashing [10, 12], PCA or LDA-based methods [13, 14], quantization  ...  Binary descriptors allow faster similarity computation than real-valued ones while requiring much less storage.  ...  values per dimension.  ... 
doi:10.1016/j.patrec.2012.08.006 fatcat:3znq765zt5gl7omss2khpy4zbm

Metric-Learning based Deep Hashing Network for Content Based Retrieval of Remote Sensing Images [article]

Subhankar Roy and Enver Sangineto and Begüm Demir and Nicu Sebe
2019 arXiv   pre-print
Experiments carried out on two benchmark RS archives point out that the proposed network significantly improves the retrieval performance under the same retrieval time when compared to the state-of-the-art  ...  Hashing methods have been recently found very effective in retrieval of remote sensing (RS) images due to their computational efficiency and fast search speed.  ...  every image in the archive based upon the image descriptors.  ... 
arXiv:1904.01258v1 fatcat:maqszcawmzfppmv74vohhe5eiu

Set-to-Set Hashing with Applications in Visual Recognition [article]

I-Hong Jhuo, Jun Wang
2019 arXiv   pre-print
For the query set process, we adopt a learning-to-hash pipeline that turns the kernel representations into hash bits based on simple learners, using multiple kernel learning.  ...  One challenge stands out in solving the problem---set representation and measure of similarity. Particularly, the query set and the sets in dataset collection can have varying cardinalities.  ...  Such performance gains confirm the value of exploring statistical and clique-based structural information for hash function design.  ... 
arXiv:1711.00888v2 fatcat:curyj2mqtfdhhh2kg6wwsovzoi

Near-Duplicate Keyframe Identification With Interest Point Matching and Pattern Learning

Wan-Lei Zhao, Chong-Wah Ngo, Hung-Khoon Tan, Xiao Wu
2007 IEEE transactions on multimedia  
By exploring the properties of PCA-SIFT, the filtering capability and speed of LIP-IS are asymptotically estimated and compared to locality sensitive hashing (LSH).  ...  The issues in matching reliability, filtering efficiency and learning flexibility are novelly exploited to delve into the potential of LIP-based retrieval and detection.  ...  During NDK ranking, the keyframe similarity is based on the cardinality of matched pairs.  ... 
doi:10.1109/tmm.2007.898928 fatcat:6qoj5vmi6bechnod5tvbn6stom

Efficient Nearest Neighbors via Robust Sparse Hashing

Anoop Cherian, Suvrit Sra, Vassilios Morellas, Nikolaos Papanikolopoulos
2014 IEEE Transactions on Image Processing  
of the hash code generated from the sparse code seldom happens, thereby breaking the NN retrieval.  ...  His transportation research has included projects involving vision-based sensing and classification of vehicles, and the human activity recognition.  ...  We are grateful for resources from the University of Minnesota Supercomputing Institute.  ... 
doi:10.1109/tip.2014.2324280 pmid:25122742 fatcat:rcopo4azqfeqdhpqjx2svdq55m

Scalable Prototype Selection by Genetic Algorithms and Hashing [article]

Yenisel Plasencia-Calaña, Mauricio Orozco-Alzate, Heydi Méndez-Vázquez, Edel García-Reyes, Robert P.W. Duin
2017 arXiv   pre-print
The methods are based on genetic algorithms, dissimilarity-based hashing, and two different unsupervised and supervised scalable criteria.  ...  The supervised criterion is based on counting matching labels of objects and their closest prototypes.  ...  Acknowledgement We acknowledge financial support from the FET programme within the EU FP7, under the SIMBAD project (contract 213250).  ... 
arXiv:1712.09277v1 fatcat:sf352s5c5je2xeqmoo43cf2wde

Discrete Multi-modal Hashing with Canonical Views for Robust Mobile Landmark Search [article]

Lei Zhu, Zi Huang, Xiaobai Liu, Xiangnan He, Jingkuan Song, Xiaofang Zhou
2017 arXiv   pre-print
One is high bandwidth consumption of query transmission, and the other is the huge visual variations of query images sent from mobile devices.  ...  Experiments on real world landmark datasets demonstrate the superior performance of CV-DMH over several state-of-the-art methods.  ...  ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers for their constructive and helpful suggestions.  ... 
arXiv:1707.04047v1 fatcat:in4v4b4i2zarblkbuj2ajba254

Multimedia semantics-aware query-adaptive hashing with bits reconfigurability

Yadong Mu, Xiangyu Chen, Xianglong Liu, Tat-Seng Chua, Shuicheng Yan
2012 International Journal of Multimedia Information Retrieval  
Extensive studies are conducted on largescale image benchmarks to comparatively investigate the performance of different strategies for reconfigurable hashing.  ...  Despite the vast literature on hashing, to our best knowledge rare endeavors have been spent toward the reusability of hashing structures in large-scale data sets.  ...  Denote the hashing vectors after refinement to be F. The key idea is akin to the idea of supervised localitypreserving method based on graph Laplacian [10].  ... 
doi:10.1007/s13735-012-0003-7 fatcat:5mzhax6kdjdyxa7y2ctyhtmne4

On security threats for robust perceptual hashing

O. Koval, S. Voloshynovskiy, P. Bas, F. Cayre, Ping Wah Wong, Edward J. Delp III, Jana Dittmann, Nasir D. Memon
2009 Media Forensics and Security  
relies on the difficulty to tamper the hash.  ...  A theoretical analysis of the information leakage in the context of Random-Based Hash is proposed.  ...  Future work should be conducted on more sophisticated content-based hashing schemes which use a more important number of descriptors 29 and on practical implementations of random-based hashing systems  ... 
doi:10.1117/12.807487 dblp:conf/mediaforensics/KovalVBC09 fatcat:4m73lfq6vzhshcq32tyxhxayl4

Robust Sparse Hashing

Anoop Cherian, Vassilios Morellas, Nikolaos Papanikolopoulos
2012 2012 19th IEEE International Conference on Image Processing  
We tackle this difficulty via our novel dictionary learning and sparse coding framework called RSH by learning dictionaries on the robustified counterparts of uncertain data points.  ...  The algorithm is applied to NN retrieval for Scale Invariant Feature Transform (SIFT) descriptors.  ...  Image pairs were chosen from each category and NN is applied on the descriptors from these pairs, where the descriptors from the first image is used as the database and the other as the query.  ... 
doi:10.1109/icip.2012.6467385 dblp:conf/icip/CherianMP12 fatcat:nmhjmukcjnbxdmskohsldvt77y

Trinary-Projection Trees for Approximate Nearest Neighbor Search

Jingdong Wang, Naiyan Wang, You Jia, Jian Li, Gang Zeng, Hongbin Zha, Xian-Sheng Hua
2014 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We address the problem of approximate nearest neighbor (ANN) search for visual descriptor indexing.  ...  We justify our approach on large scale local patch indexing and similar image search.  ...  Local feature-based object retrieval methods include the step of searching a huge database of patch descriptors for most similar descriptors [38] .  ... 
doi:10.1109/tpami.2013.125 pmid:24356357 fatcat:dawzydol5na7be55fmcd3tfuse
« Previous Showing results 1 — 15 out of 276 results