This page has only limited features, please log in for full access.

Unclaimed
Chengyuan Zhang
College of Computer Science and Electronic Engineering, Hunan University, Changsha, People’s Republic of China

Honors and Awards

The user has no records in this section


Career Timeline

The user has no records in this section.


Short Biography

The user biography is not available.
Following
Followers
Co Authors
The list of users this user is following is empty.
Following: 0 users

Feed

Article
Published: 18 June 2021 in Neural Processing Letters
Reads 0
Downloads 0

Recently, massive multimedia data (especially images) is moved to the cloud environment for analysis and retrieval, which makes data security issue become particularly significant. Image similarity join has attracted more and more attention in the community of multimedia retrieval. However, few researches have investigated the privacy-preserving problem of image similarity join. To tackle this challenge, this paper proposes a novel privacy-preserving image similarity join method, called PPIS-JOIN. Different from the existing schemes, this approach aims to combine deep image hashing method and a novel affine transformation method to conceal sensitive information at feature level and generate high quality hash codes. Meanwhile, based on secure hash codes, a privacy-preserving similarity query model is proposed, which includes a secure image hash codes based inverted index, called ISH-Index, to support efficient and accuracy similarity search. We conduct comprehensive experiments on three common used benchmarks, and the results demonstrate the performance of the proposed PPIS-JOIN outperforms baselines.

ACS Style

Chengyuan Zhang; Fangxin Xie; Hao Yu; Jianfeng Zhang; Lei Zhu; Yangding Li. PPIS-JOIN: A Novel Privacy-Preserving Image Similarity Join Method. Neural Processing Letters 2021, 1 -19.

AMA Style

Chengyuan Zhang, Fangxin Xie, Hao Yu, Jianfeng Zhang, Lei Zhu, Yangding Li. PPIS-JOIN: A Novel Privacy-Preserving Image Similarity Join Method. Neural Processing Letters. 2021; ():1-19.

Chicago/Turabian Style

Chengyuan Zhang; Fangxin Xie; Hao Yu; Jianfeng Zhang; Lei Zhu; Yangding Li. 2021. "PPIS-JOIN: A Novel Privacy-Preserving Image Similarity Join Method." Neural Processing Letters , no. : 1-19.

Article
Published: 07 March 2021 in Neural Processing Letters
Reads 0
Downloads 0

Privacy-preserving cross-modal retrieval is a significant problem in the area of multimedia analysis. As the amount of data is exploding, cross-modal data analysis and retrieval is often realized on cloud computing environment. Therefore, the privacy protection of large-scale cross-modal data has become a problem that can not be ignored. To further improve the accuracy and efficiency of privacy-preserving search, this paper proposes a novel cross-modal hashing scheme, named deep adversarial privacy-preserving cross-modal hashing (DAP\(^2\)CMH). This method consists of a deep cross-modal hashing model termed DACMH, and a secure index structure called CMH\(^2\)-Tree. The former is a combination of deep hashing and adversarial learning to capture intra-modal and inter-modal correlation. The latter is a hierarchical hashing index structure that can provide efficient data organization based on cross-modal hash codes. We conduct comprehensive experiments on three common used benchmarks. The results show that the proposed approach DAP\(^2\)CMH outperforms the state-of-the-arts.

ACS Style

Lei Zhu; Jiayu Song; Zhan Yang; Wenti Huang; Chengyuan Zhang; Weiren Yu. DAP$$^2$$CMH: Deep Adversarial Privacy-Preserving Cross-Modal Hashing. Neural Processing Letters 2021, 1 -21.

AMA Style

Lei Zhu, Jiayu Song, Zhan Yang, Wenti Huang, Chengyuan Zhang, Weiren Yu. DAP$$^2$$CMH: Deep Adversarial Privacy-Preserving Cross-Modal Hashing. Neural Processing Letters. 2021; ():1-21.

Chicago/Turabian Style

Lei Zhu; Jiayu Song; Zhan Yang; Wenti Huang; Chengyuan Zhang; Weiren Yu. 2021. "DAP$$^2$$CMH: Deep Adversarial Privacy-Preserving Cross-Modal Hashing." Neural Processing Letters , no. : 1-21.

Journal article
Published: 11 August 2020 in IEEE MultiMedia
Reads 0
Downloads 0

Cross-modal retrieval has become a hot issue in this years. Many existing works pay attentions on correlation learning to generate a common subspace for cross-modal correlation measurement, and others uses adversarial learning technique to abate the heterogeneity of multi-modal data. However, very few works combine correlation learning and adversarial learning to bridge the inter-modal semantic gap and diminish cross-modal heterogeneity. This paper proposes a novel cross-modal retrieval method, named ALSCOR, which is an end-to-end framework to integrate cross-modal representation learning, correlation learning and adversarial. CCA model, combined with VisNet and TxtNet, is proposed to capture cross-modal non-linear correlation. Besides, intra-modal classifier and modality classifier are used to learn intra-modal discrimination and minimize the inter-modal heterogeneity. Comprehensive experiments are conducted on three benchmark datasets. The results demonstrate that the proposed ALSCOR has better performance than the state-of-the-arts.

ACS Style

Lei Zhu; Jiayu Song; Xiaofeng Zhu; Chengyuan Zhang; Shichao Zhang; Xinpan Yuan; Yang Wang. Adversarial Learning-Based Semantic Correlation Representation for Cross-Modal Retrieval. IEEE MultiMedia 2020, 27, 79 -90.

AMA Style

Lei Zhu, Jiayu Song, Xiaofeng Zhu, Chengyuan Zhang, Shichao Zhang, Xinpan Yuan, Yang Wang. Adversarial Learning-Based Semantic Correlation Representation for Cross-Modal Retrieval. IEEE MultiMedia. 2020; 27 (4):79-90.

Chicago/Turabian Style

Lei Zhu; Jiayu Song; Xiaofeng Zhu; Chengyuan Zhang; Shichao Zhang; Xinpan Yuan; Yang Wang. 2020. "Adversarial Learning-Based Semantic Correlation Representation for Cross-Modal Retrieval." IEEE MultiMedia 27, no. 4: 79-90.

Journal article
Published: 12 April 2020 in Neurocomputing
Reads 0
Downloads 0

With the rapid development of multimedia techniques, mobile Internet and cloud computing, large-scale image retrieval service has become a necessity in our daily life. To overcome the challenges of privacy-preserving and data security during the image retrieval, we propose a novel deep hashing based privacy-preserving image retrieval method named TDHPPIR that can generate high quality hash codes of image and provide an efficient index structure for fast image retrieval in a security manner in cloud. A triplet Deep CNN structure model is introduced to learn deep visual representations and hash codes of images simultaneously, which can generate higher quality hash codes than the traditional way that mainly utilizes hand-crafted features. Besides, A novel hierachical bit-scalable hash codes based S-Tree, named H2S-Tree, is developed to increase the search efficiency. Comprehensive experiments on four benchmarks show that our method can combat the state-of-the-arts in both aspects of accuracy and efficiency.

ACS Style

Chengyuan Zhang; Lei Zhu; Shichao Zhang; Weiren Yu. TDHPPIR: An Efficient Deep Hashing Based Privacy-Preserving Image Retrieval Method. Neurocomputing 2020, 406, 386 -398.

AMA Style

Chengyuan Zhang, Lei Zhu, Shichao Zhang, Weiren Yu. TDHPPIR: An Efficient Deep Hashing Based Privacy-Preserving Image Retrieval Method. Neurocomputing. 2020; 406 ():386-398.

Chicago/Turabian Style

Chengyuan Zhang; Lei Zhu; Shichao Zhang; Weiren Yu. 2020. "TDHPPIR: An Efficient Deep Hashing Based Privacy-Preserving Image Retrieval Method." Neurocomputing 406, no. : 386-398.

Journal article
Published: 23 January 2020 in IEEE Access
Reads 0
Downloads 0
ACS Style

Lei Zhu; Jiayu Song; Weiren Yu; Chengyuan Zhang; Hao Yu; Zuping Zhang. Reverse Spatial Visual Top-$k$ Query. IEEE Access 2020, 8, 21770 -21787.

AMA Style

Lei Zhu, Jiayu Song, Weiren Yu, Chengyuan Zhang, Hao Yu, Zuping Zhang. Reverse Spatial Visual Top-$k$ Query. IEEE Access. 2020; 8 ():21770-21787.

Chicago/Turabian Style

Lei Zhu; Jiayu Song; Weiren Yu; Chengyuan Zhang; Hao Yu; Zuping Zhang. 2020. "Reverse Spatial Visual Top-$k$ Query." IEEE Access 8, no. : 21770-21787.

Journal article
Published: 27 December 2019 in Neurocomputing
Reads 0
Downloads 0

Person re-identification (person Re-Id) aims to retrieve the pedestrian images of the same person that captured by disjoint and non-overlapping cameras. Lots of researchers recently focused on this hot issue and proposed deep learning based methods to enhance the recognition rate in a supervised or unsupervised manner. However,there are two limitations that cannot be ignored: firstly, compared with other image retrieval benchmarks, the size of existing person Re-Id datasets is far from meeting the requirement, which cannot provide sufficient pedestrian samples for the training of deep model; secondly, the samples in existing datasets do not have sufficient human motions or postures coverage to provide more priori knowledges for learning. In this paper, we introduce a novel unsupervised pose augmentation cross-view person Re-Id scheme called PAC-GAN to overcome these limitations. We firstly present the formal definition of cross-view pose augmentation and then propose the framework of PAC-GAN that is a novel conditional generative adversarial network (CGAN) based approach to improve the performance of unsupervised corss-view person Re-Id. Specifically, the pose generation model in PAC-GAN called CPG-Net is to generate enough quantity of pose-rich samples from original image and skeleton samples. The pose augmentation dataset is produced by combining the synthesized pose-rich samples with the original samples, which is fed into the corss-view person Re-Id model named Cross-GAN. Besides, we use weight-sharing strategy in the CPG-Net to improve the quality of new generated samples. To the best of our knowledge, we are the first to enhance the unsupervised cross-view person Re-Id by pose augmentation, and the results of extensive experiments show that the proposed scheme can combat the state-of-the-arts with recognition rate.

ACS Style

Chengyuan Zhang; Lei Zhu; Shichao Zhang; Weiren Yu. PAC-GAN: An effective pose augmentation scheme for unsupervised cross-view person re-identification. Neurocomputing 2019, 387, 22 -39.

AMA Style

Chengyuan Zhang, Lei Zhu, Shichao Zhang, Weiren Yu. PAC-GAN: An effective pose augmentation scheme for unsupervised cross-view person re-identification. Neurocomputing. 2019; 387 ():22-39.

Chicago/Turabian Style

Chengyuan Zhang; Lei Zhu; Shichao Zhang; Weiren Yu. 2019. "PAC-GAN: An effective pose augmentation scheme for unsupervised cross-view person re-identification." Neurocomputing 387, no. : 22-39.

Journal article
Published: 21 October 2019 in IEEE Access
Reads 0
Downloads 0

In the big data era, massive amount of multimedia data with geo-tags has been generated and collected by smart devices equipped with mobile communications module and position sensor module. This trend has put forward higher request on large-scale geo-multimedia retrieval. Spatial similarity join is one of the significant problems in the area of spatial database. Previous works focused on spatial textual document search problem, rather than geo-multimedia retrieval. In this paper, we investigate a novel geo-multimedia retrieval paradigm named spatial visual similarity join (SVS-JOIN for short), which aims to search similar geo-image pairs in both aspects of geo-location and visual content. Firstly, the definition of SVS-JOIN is proposed and then we present the geographical similarity and visual similarity measurement. Inspired by the approach for textual similarity join, we develop an algorithm named SVS-JOINB by combining the PPJOIN algorithm and visual similarity. Besides, an extension of it named SVS-JOING is developed, which utilizes spatial grid strategy to improve the search efficiency. To further speed up the search, a novel approach called SVS-JOINQ is carefully designed, in which a quadtree and a global inverted index are employed. Comprehensive experiments are conducted on two geo-image datasets and the results demonstrate that our solution can address the SVS-JOIN problem effectively and efficiently.

ACS Style

Lei Zhu; Weiren Yu; Chengyuan Zhang; Zuping Zhang; Fang Huang; Hao Yu. SVS-JOIN: Efficient Spatial Visual Similarity Join for Geo-Multimedia. IEEE Access 2019, 7, 158389 -158408.

AMA Style

Lei Zhu, Weiren Yu, Chengyuan Zhang, Zuping Zhang, Fang Huang, Hao Yu. SVS-JOIN: Efficient Spatial Visual Similarity Join for Geo-Multimedia. IEEE Access. 2019; 7 (99):158389-158408.

Chicago/Turabian Style

Lei Zhu; Weiren Yu; Chengyuan Zhang; Zuping Zhang; Fang Huang; Hao Yu. 2019. "SVS-JOIN: Efficient Spatial Visual Similarity Join for Geo-Multimedia." IEEE Access 7, no. 99: 158389-158408.

Journal article
Published: 09 September 2019 in IEEE Access
Reads 0
Downloads 0

Due to the rapid development of mobile Internet techniques, such as online social networking and location-based services, massive amount of multimedia data with geographical information is generated and uploaded to the Internet. In this paper, we propose a novel type of cross-modal multimedia retrieval, called geo-multimedia cross-modal retrieval, which aims to find a set of geo-multimedia objects according to geographical distance proximity and semantic concept similarity. Previous studies for cross-modal retrieval and spatial keyword search cannot address this problem effectively because they do not consider multimedia data with geo-tags (geo-multimedia). Firstly, we present the definition of kNN geo-multimedia cross-modal query and introduce relevant concepts such as spatial distance and semantic similarity measurement. As the key notion of this work, cross-modal semantic representation space is formulated at the first time. A novel framework for geo-multimedia cross-modal retrieval is proposed, which includes multi-modal feature extraction, cross-modal semantic space mapping, geo-multimedia spatial index and cross-modal semantic similarity measurement. To bridge the semantic gap between different modalities, we also propose a method named cross-modal semantic matching (CoSMat for shot) which contains two important components, i.e., CorrProj and LogsTran, which aims to build a common semantic representation space for cross-modal semantic similarity measurement. In addition, to implement semantic similarity measurement, we employ deep learning based method to learn multi-modal features that contains more high level semantic information. Moreover, a novel hybrid index, GMR-Tree is carefully designed, which combines signatures of semantic representations and R-Tree. An efficient GMR-Tree based kNN search algorithm called kGMCMS is developed. Comprehensive experimental evaluations on real and synthetic datasets clearly demonstrate that our approach outperforms the-state-of-the-art methods.

ACS Style

Lei Zhu; Jun Long; Chengyuan Zhang; Weiren Yu; Xinpan Yuan; Longzhi Sun. An Efficient Approach for Geo-Multimedia Cross-Modal Retrieval. IEEE Access 2019, 7, 180571 -180589.

AMA Style

Lei Zhu, Jun Long, Chengyuan Zhang, Weiren Yu, Xinpan Yuan, Longzhi Sun. An Efficient Approach for Geo-Multimedia Cross-Modal Retrieval. IEEE Access. 2019; 7 (99):180571-180589.

Chicago/Turabian Style

Lei Zhu; Jun Long; Chengyuan Zhang; Weiren Yu; Xinpan Yuan; Longzhi Sun. 2019. "An Efficient Approach for Geo-Multimedia Cross-Modal Retrieval." IEEE Access 7, no. 99: 180571-180589.

Journal article
Published: 07 February 2019 in Neurocomputing
Reads 0
Downloads 0

Person re-identification (re-id) refers to matching pedestrians across disjoint yet non-overlapping camera views. The most effective way to match these pedestrians undertaking significant visual variations is to seek reliably invariant features that can describe the person of interest faithfully. Most of existing methods are presented in a supervised manner to produce discriminative features by relying on labeled paired images in correspondence. However, annotating pair-wise images is prohibitively expensive in labours, and thus not practical in large-scale networked cameras. Moreover, seeking comparable representations across camera views demands a flexible model to address the complex distributions of images. In this work, we study the co-occurrence statistic patterns between pairs of images, and propose to crossing Generative Adversarial Network (Cross-GAN) for learning a joint distribution for cross-image representations in an unsupervised manner. Given a pair of person images, the proposed model consists of the variational auto-encoder to encode the pair into respective latent variables, a proposed cross-view alignment to reduce the view disparity, and an adversarial layer to seek the joint distribution of latent representations. The learned latent representations are well-aligned to reflect the co-occurrence patterns of paired images. We empirically evaluate the proposed model against challenging datasets, and our results show the importance of joint invariant features in improving matching rates of person re-id with comparison to semi/unsupervised state-of-the-arts.

ACS Style

Chengyuan Zhang; Lin Wu; Yang Wang. Crossing generative adversarial networks for cross-view person re-identification. Neurocomputing 2019, 340, 259 -269.

AMA Style

Chengyuan Zhang, Lin Wu, Yang Wang. Crossing generative adversarial networks for cross-view person re-identification. Neurocomputing. 2019; 340 ():259-269.

Chicago/Turabian Style

Chengyuan Zhang; Lin Wu; Yang Wang. 2019. "Crossing generative adversarial networks for cross-view person re-identification." Neurocomputing 340, no. : 259-269.

Conference paper
Published: 21 November 2018 in Privacy Enhancing Technologies
Reads 0
Downloads 0

Massive amount of data that are geo-tagged and associated with text information are being generated at an unprecedented scale in many emerging applications such as location based services and social networks. Due to their importance, a large body of work has focused on efficiently computing various spatial keyword queries. In this paper, we study the top-k temporal spatial keyword query which considers three important constraints during the search including time, spatial proximity and textual relevance. A novel index structure, namely SSG-tree, to efficiently insert/delete spatio-temporal web objects with high rates. Base on SSG-tree an efficient algorithm is developed to support top-k temporal spatial keyword query. We show via extensive experimentation with real spatial databases that our method has increased performance over alternate techniques.

ACS Style

Chengyuan Zhang; Lei Zhu; Weiren Yu; Jun Long; Fang Huang; Hongbo Zhao. Efficient Top K Temporal Spatial Keyword Search. Privacy Enhancing Technologies 2018, 80 -92.

AMA Style

Chengyuan Zhang, Lei Zhu, Weiren Yu, Jun Long, Fang Huang, Hongbo Zhao. Efficient Top K Temporal Spatial Keyword Search. Privacy Enhancing Technologies. 2018; ():80-92.

Chicago/Turabian Style

Chengyuan Zhang; Lei Zhu; Weiren Yu; Jun Long; Fang Huang; Hongbo Zhao. 2018. "Efficient Top K Temporal Spatial Keyword Search." Privacy Enhancing Technologies , no. : 80-92.

Conference paper
Published: 21 November 2018 in Privacy Enhancing Technologies
Reads 0
Downloads 0

With advances in geo-positioning technologies and geo-locat-ion services, there are a rapidly growing massive amount of spatio-tempor-al data collected in many applications such as location-aware devices and wireless communication, in which an object is described by its spatial location and its timestamp. Consequently, the study of spatio-temporal search which explores both geo-location information and temporal information of the data has attracted significant concern from research organizations and commercial communities. This work study the problem of spatio-temporal k-nearest neighbors search (STkNNS), which is fundamental in the spatial temporal queries. Based on HBase, a novel index structure is proposed, called Hybrid Spatio-Temporal HBase Index (HSTI for short), which is carefully designed and takes both spatial and temporal information into consideration to effectively reduce the search space. Based on HSTI, an efficient algorithm is developed to deal with spatio-temporal k-nearest neighbors search. Comprehensive experiments on real and synthetic data clearly show that HSTI is three to five times faster than the state-of-the-art technique.

ACS Style

Chengyuan Zhang; Lei Zhu; Jun Long; Shuangqiao Lin; Zhan Yang; Wenti Huang. A Hybrid Index Model for Efficient Spatio-Temporal Search in HBase. Privacy Enhancing Technologies 2018, 108 -120.

AMA Style

Chengyuan Zhang, Lei Zhu, Jun Long, Shuangqiao Lin, Zhan Yang, Wenti Huang. A Hybrid Index Model for Efficient Spatio-Temporal Search in HBase. Privacy Enhancing Technologies. 2018; ():108-120.

Chicago/Turabian Style

Chengyuan Zhang; Lei Zhu; Jun Long; Shuangqiao Lin; Zhan Yang; Wenti Huang. 2018. "A Hybrid Index Model for Efficient Spatio-Temporal Search in HBase." Privacy Enhancing Technologies , no. : 108-120.

Conference paper
Published: 21 November 2018 in Privacy Enhancing Technologies
Reads 0
Downloads 0

With the rapid development of mobile computing and Web services, a huge amount of data with spatial and temporal information have been collected everyday by smart mobile terminals, in which an object is described by its spatial information and temporal information. Motivated by the significance of spatio-temporal range search and the lack of efficient search algorithm, in this paper, we study the problem of spatio-temporal range search (STRS), a novel index structure is proposed, called HOC-Tree, which is based on Hilbert curve and OC-Tree, and takes both spatial and temporal information into consideration. Based on HOC-Tree, we develop an efficient algorithm to solve the problem of spatio-temporal range search. Comprehensive experiments on real and synthetic data demonstrate that our method is more efficient than the state-of-the-art technique.

ACS Style

Jun Long; Lei Zhu; Chengyuan Zhang; Shuangqiao Lin; Zhan Yang; Xinpan Yuan. HOC-Tree: A Novel Index for Efficient Spatio-Temporal Range Search. Privacy Enhancing Technologies 2018, 93 -107.

AMA Style

Jun Long, Lei Zhu, Chengyuan Zhang, Shuangqiao Lin, Zhan Yang, Xinpan Yuan. HOC-Tree: A Novel Index for Efficient Spatio-Temporal Range Search. Privacy Enhancing Technologies. 2018; ():93-107.

Chicago/Turabian Style

Jun Long; Lei Zhu; Chengyuan Zhang; Shuangqiao Lin; Zhan Yang; Xinpan Yuan. 2018. "HOC-Tree: A Novel Index for Efficient Spatio-Temporal Range Search." Privacy Enhancing Technologies , no. : 93-107.

Article
Published: 31 October 2018 in Multimedia Tools and Applications
Reads 0
Downloads 0

With the proliferation of online social networking services and mobile smart devices equipped with mobile communications module and position sensor module, massive amount of multimedia data has been collected, stored and shared. This trend has put forward higher request on massive multimedia data retrieval. In this paper, we investigate a novel spatial query named region of visual interests query (RoVIQ), which aims to search users containing geographical information and visual words. Three baseline methods are presented to introduce how to exploit existing techniques to address this problem. Then we propose the definition of this query and related notions at the first time. To improve the performance of query, we propose a novel spatial indexing structure called quadtree based inverted visual index which is a combination of quadtree, inverted index and visual words. Based on it, we design a efficient search algorithm named region of visual interests search to support RoVIQ. Experimental evaluations on real geo-image datasets demonstrate that our solution outperforms state-of-the-art method.

ACS Style

Chengyuan Zhang; YunWu Lin; Lei Zhu; Zuping Zhang; Yan Tang; Fang Huang. Efficient region of visual interests search for geo-multimedia data. Multimedia Tools and Applications 2018, 78, 30839 -30863.

AMA Style

Chengyuan Zhang, YunWu Lin, Lei Zhu, Zuping Zhang, Yan Tang, Fang Huang. Efficient region of visual interests search for geo-multimedia data. Multimedia Tools and Applications. 2018; 78 (21):30839-30863.

Chicago/Turabian Style

Chengyuan Zhang; YunWu Lin; Lei Zhu; Zuping Zhang; Yan Tang; Fang Huang. 2018. "Efficient region of visual interests search for geo-multimedia data." Multimedia Tools and Applications 78, no. 21: 30839-30863.

Preprint
Published: 14 October 2018
Reads 0
Downloads 0

This paper aims to solve the problem of large-scale video retrieval by a query image. Firstly, we define the problem of top-$k$ image to video query. Then, we combine the merits of convolutional neural networks(CNN for short) and Bag of Visual Word(BoVW for short) module to design a model for video frames information extraction and representation. In order to meet the requirements of large-scale video retrieval, we proposed a visual weighted inverted index(VWII for short) and related algorithm to improve the efficiency and accuracy of retrieval process. Comprehensive experiments show that our proposed technique achieves substantial improvements (up to an order of magnitude speed up) over the state-of-the-art techniques with similar accuracy.

ACS Style

Chengyuan Zhang; YunWu Lin; Lei Zhu; Anfeng Liu; Zuping Zhang; Fang Huang. CNN-VWII: An Efficient Approach for Large-Scale Video Retrieval by Image Queries. 2018, 1 .

AMA Style

Chengyuan Zhang, YunWu Lin, Lei Zhu, Anfeng Liu, Zuping Zhang, Fang Huang. CNN-VWII: An Efficient Approach for Large-Scale Video Retrieval by Image Queries. . 2018; ():1.

Chicago/Turabian Style

Chengyuan Zhang; YunWu Lin; Lei Zhu; Anfeng Liu; Zuping Zhang; Fang Huang. 2018. "CNN-VWII: An Efficient Approach for Large-Scale Video Retrieval by Image Queries." , no. : 1.

Article
Published: 02 October 2018 in Multimedia Tools and Applications
Reads 0
Downloads 0

With the rapid development of mobile Internet and cloud computing technology, large-scale multimedia data, e.g., texts, images, audio and videos have been generated, collected, stored and shared. In this paper, we propose a novel query problem named continuous top-k geo-image query on road network which aims to search out a set of geo-visual objects based on road network distance proximity and visual content similarity. Existing approaches for spatial textual query and geo-image query cannot address this problem effectively because they do not consider both of visual content similarity and road network distance proximity on road network. In order to address this challenge effectively and efficiently, firstly we propose the definition of geo-visual objects and continuous top-k geo-visual objects query on road network, then develop a score function for search. To improve the query efficiency in a large-scale road network, we propose the search algorithm named geo-visual search on road network based on a novel hybrid indexing framework called VIG-Tree, which combines G-Tree and visual inverted index technique. In addition, an important notion named safe interval and results updating rule are proposed, and based on them we develop an efficient algorithm named moving monitor algorithm to solve continuous query. Experimental evaluation on real multimedia dataset and road network dataset illustrates that our solution outperforms state-of-the-art method.

ACS Style

Chengyuan Zhang; Kesheng Cheng; Lei Zhu; Ruipeng Chen; Zuping Zhang; Fang Huang. Efficient continuous top-k geo-image search on road network. Multimedia Tools and Applications 2018, 78, 30809 -30838.

AMA Style

Chengyuan Zhang, Kesheng Cheng, Lei Zhu, Ruipeng Chen, Zuping Zhang, Fang Huang. Efficient continuous top-k geo-image search on road network. Multimedia Tools and Applications. 2018; 78 (21):30809-30838.

Chicago/Turabian Style

Chengyuan Zhang; Kesheng Cheng; Lei Zhu; Ruipeng Chen; Zuping Zhang; Fang Huang. 2018. "Efficient continuous top-k geo-image search on road network." Multimedia Tools and Applications 78, no. 21: 30809-30838.

Journal article
Published: 01 October 2018 in IEEE Access
Reads 0
Downloads 0

Deep convolutional neural networks (DCNNs) are currently popular in human activity recognition (HAR) applications. However, in the face of modern artificial intelligence sensor-based games, many research achievements cannot be practically applied on portable devices (i.e., smart phone, VR/AR). DCNNs are typically resource-intensive and too large to be deployed on portable devices, and thus, this limits the practical application of complex activity detection. In addition, since portable devices do not possess high-performance graphic processing units, there is hardly any improvement in Action Game (ACT) experience. Besides, in order to deal with multi-sensor collaboration, all previous HAR models typically treated the representations from different sensor signal sources equally. However, distinct types of activities should adopt different fusion strategies. In this paper, a novel scheme is proposed. This scheme is used to train 2-bit CNNs with weights and activations constrained to {−0.5, 0, 0.5}. It takes into account the correlation between different sensor signal sources and the activity types. This model, which we refer to as DFTerNet, aims at producing a more reliable inference and better trade-offs for practical applications. It is known that quantization of weights and activations can substantially reduce memory size and use more efficient bitwise operations to replace floating or matrix operations to achieve much faster calculation and lower power consumption. Our basic idea is to exploit quantization of weights and activations directly in pre-trained filter banks and adopt dynamic fusion strategies for different activity types. Experiments demonstrate that by using a dynamic fusion strategy, it is possible to exceed the baseline model performance by up to ~5% on activity recognition data sets, such as the OPPORTUNITY and PAMAP2 data sets. Using the quantization method proposed, we were able to achieve performances closer to that of the full-precision counterpart. These results were also verified using the UniMiB-SHAR data set. In addition, the proposed method can achieve $\sim 9\times $ acceleration on CPUs and $\sim 11\times $ memory saving.

ACS Style

Zhan Yang; Osolo Ian Raymond; Chengyuan Zhang; Ying Wan; Jun Long. DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition. IEEE Access 2018, 6, 56750 -56764.

AMA Style

Zhan Yang, Osolo Ian Raymond, Chengyuan Zhang, Ying Wan, Jun Long. DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition. IEEE Access. 2018; 6 ():56750-56764.

Chicago/Turabian Style

Zhan Yang; Osolo Ian Raymond; Chengyuan Zhang; Ying Wan; Jun Long. 2018. "DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition." IEEE Access 6, no. : 56750-56764.

Journal article
Published: 20 September 2018 in Journal of Advanced Computational Intelligence and Intelligent Informatics
Reads 0
Downloads 0

Image similarity measures play an important role in nearest neighbor search and duplicate detection for large-scale image datasets. Recently, Minwise Hashing (or Minhash) and its related hashing algorithms have achieved great performances in large-scale image retrieval systems. However, there are a large number of comparisons for image pairs in these applications, which may spend a lot of computation time and affect the performance. In order to quickly obtain the pairwise images that theirs similarities are higher than the specific thresholdT(e.g., 0.5), we propose a dynamic threshold filter of Minwise Hashing for image similarity measures. It greatly reduces the calculation time by terminating the unnecessary comparisons in advance. We also find that the filter can be extended to other hashing algorithms, on when the estimator satisfies the binomial distribution, such as b-Bit Minwise Hashing, One Permutation Hashing, etc. In this pager, we use the Bag-of-Visual-Words (BoVW) model based on the Scale Invariant Feature Transform (SIFT) to represent the image features. We have proved that the filter is correct and effective through the experiment on real image datasets.

ACS Style

Jun Long; Qunfeng Liu; Xinpan Yuan; Chengyuan Zhang; Junfeng Liu. A Filter of Minhash for Image Similarity Measures. Journal of Advanced Computational Intelligence and Intelligent Informatics 2018, 22, 689 -698.

AMA Style

Jun Long, Qunfeng Liu, Xinpan Yuan, Chengyuan Zhang, Junfeng Liu. A Filter of Minhash for Image Similarity Measures. Journal of Advanced Computational Intelligence and Intelligent Informatics. 2018; 22 (5):689-698.

Chicago/Turabian Style

Jun Long; Qunfeng Liu; Xinpan Yuan; Chengyuan Zhang; Junfeng Liu. 2018. "A Filter of Minhash for Image Similarity Measures." Journal of Advanced Computational Intelligence and Intelligent Informatics 22, no. 5: 689-698.

Preprint
Published: 08 September 2018
Reads 0
Downloads 0
ACS Style

Chengyuan Zhang; YunWu Lin; Lei Zhu; Zuping Zhang; Xinpan Yuan; Fang Huang. Efficient Multimedia Similarity Measurement Using Similar Elements. 2018, 1 .

AMA Style

Chengyuan Zhang, YunWu Lin, Lei Zhu, Zuping Zhang, Xinpan Yuan, Fang Huang. Efficient Multimedia Similarity Measurement Using Similar Elements. . 2018; ():1.

Chicago/Turabian Style

Chengyuan Zhang; YunWu Lin; Lei Zhu; Zuping Zhang; Xinpan Yuan; Fang Huang. 2018. "Efficient Multimedia Similarity Measurement Using Similar Elements." , no. : 1.

Journal article
Published: 31 August 2018 in Sensors
Reads 0
Downloads 0

In energy harvesting wireless sensor networks (EHWSNs), the energy tension of the network can be relieved by obtaining the energy from the surrounding environment, but the cost on hardware cannot be ignored. Therefore, how to minimize the cost of energy harvesting hardware to reduce the network deployment cost, and further optimize the network performance, is still a challenging issue in EHWSNs. In this paper, an energy conserving and transmission radius adaptive (ECTRA) scheme is proposed to reduce the cost and optimize the performance of solar-based EHWSNs. There are two main innovations of the ECTRA scheme. Firstly, an energy conserving approach is proposed to conserve energy and avoid outage for the nodes in hotspots, which are the bottleneck of the whole network. The novelty of this scheme is adaptively rotating the transmission radius. In this way, the nodes with maximum energy consumption are rotated, balancing energy consumption between nodes and reducing the maximum energy consumption in the network. Therefore, the battery storage capacity of nodes and the cost on hardware. Secondly, the ECTRA scheme selects a larger transmission radius for rotation when the node can absorb enough energy from the surroundings. The advantages of using this method are: (a) reducing the energy consumption of nodes in near-sink areas, thereby reducing the maximum energy consumption and allowing the node of the hotspot area to conserve energy, in order to prevent the node from outage. Hence, the network deployment costs can be further reduced; (b) reducing the network delay. When a larger transmission radius is used to transmit data in the network, fewer hops are needed by data packet to the sink. After the theoretical analyses, the results show the following advantages compared with traditional method. Firstly, the ECTRA scheme can effectively reduce deployment costs by 29.58% without effecting the network performance as shown in experiment analysis; Secondly, the ECTRA scheme can effectively reduce network data transmission delay by 44–71%; Thirdly, the ECTRA scheme shows a better balance in energy consumption and the maximum energy consumption is reduced by 27.89%; And lastly, the energy utilization rate is effectively improved by 30.09–55.48%.

ACS Style

Xin Ju; Wei Liu; Chengyuan Zhang; Anfeng Liu; Tian Wang; Neal N. Xiong; Zhiping Cai. An Energy Conserving and Transmission Radius Adaptive Scheme to Optimize Performance of Energy Harvesting Sensor Networks. Sensors 2018, 18, 2885 .

AMA Style

Xin Ju, Wei Liu, Chengyuan Zhang, Anfeng Liu, Tian Wang, Neal N. Xiong, Zhiping Cai. An Energy Conserving and Transmission Radius Adaptive Scheme to Optimize Performance of Energy Harvesting Sensor Networks. Sensors. 2018; 18 (9):2885.

Chicago/Turabian Style

Xin Ju; Wei Liu; Chengyuan Zhang; Anfeng Liu; Tian Wang; Neal N. Xiong; Zhiping Cai. 2018. "An Energy Conserving and Transmission Radius Adaptive Scheme to Optimize Performance of Energy Harvesting Sensor Networks." Sensors 18, no. 9: 2885.

Article
Published: 29 August 2018 in Multimedia Tools and Applications
Reads 0
Downloads 0

Due to the advances in mobile computing and multimedia techniques, there are vast amount of multimedia data with geographical information collected in multifarious applications. In this paper, we propose a novel type of image search namedinteractive geo-tagged image search which aims to find out a set of images based on geographical proximity and similarity of visual content, as well as the preference of users. Existing approaches for spatial keyword query and geo-image query cannot address this problem effectively since they do not consider these three type of information together for query. In order to solve this challenge efficiently, we propose the definition of interactive top-k geo-tagged image query and then present a framework including candidate search stage , interaction stage and termination stage. To enhance the searching efficiency in a large-scale database, we propose the candidate search algorithm named GI-SUPER Search based on a new notion called superior relationship and GIR-Tree, a novel index structure. Furthermore, two candidate selection methods are proposed for learning the preferences of the user during the interaction. At last, the termination procedure and estimation procedure are introduced in brief. Experimental evaluation on real multimedia dataset demonstrates that our solution has a really high performance.

ACS Style

Jun Long; Lei Zhu; Chengyuan Zhang; Zhan Yang; YunWu Lin; Ruipeng Chen. Efficient interactive search for geo-tagged multimedia data. Multimedia Tools and Applications 2018, 78, 30677 -30706.

AMA Style

Jun Long, Lei Zhu, Chengyuan Zhang, Zhan Yang, YunWu Lin, Ruipeng Chen. Efficient interactive search for geo-tagged multimedia data. Multimedia Tools and Applications. 2018; 78 (21):30677-30706.

Chicago/Turabian Style

Jun Long; Lei Zhu; Chengyuan Zhang; Zhan Yang; YunWu Lin; Ruipeng Chen. 2018. "Efficient interactive search for geo-tagged multimedia data." Multimedia Tools and Applications 78, no. 21: 30677-30706.