发布求助
文献互助 智能选刊 最新文献

2017 2nd International Conference on Multimedia and Image Processing (ICMIP)最新文献

筛选
英文 中文
Atrous Faster R-CNN for Small Scale Object Detection 小尺度目标检测的更快R-CNN
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.37
Tongfan Guan, Hao Zhu
{"title":"Atrous Faster R-CNN for Small Scale Object Detection","authors":"Tongfan Guan, Hao Zhu","doi":"10.1109/ICMIP.2017.37","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.37","url":null,"abstract":"Deep Convolutional Neural Networks based object detection has made significant progress recent years. However, detecting small scale objects is still a challenging task. This paper addresses the problem and proposes a unified deep neural network building upon the prominent Faster R-CNN framework. This paper has two main contributions. Firstly, an Atrous Region Proposal Network (ARPN) is proposed to explore object contexts at multiple scales by sliding a set of atrous filters with increasing dilation rates over the last convolutional feature map. Secondly, to enrich the representations of small scale image regions, this paper incorporates atrous convolution into Fast R-CNN and proposes a Dense Fast R-CNN (DFRCN), that improves the resolution of the ROI-pooled convolutional feature maps without increasing the number of parameters. In combination of the two, this paper proposes a unified network termed as Atrous Faster R-CNN. On PASCAL object detection challenge dataset, our method achieves superior performance to the state of the arts, especially for small scale objects.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125301698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 19
Counting Crowd with Fully Convolutional Networks 用全卷积网络计算人群
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.25
Jianyong Wang, Lu Wang, Fenglei Yang
{"title":"Counting Crowd with Fully Convolutional Networks","authors":"Jianyong Wang, Lu Wang, Fenglei Yang","doi":"10.1109/ICMIP.2017.25","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.25","url":null,"abstract":"Crowd counting is useful and widely applied invideo surveillance. It remains a challenge task for thecharacteristics of crowd, such as severe occlusions, sceneperspective distortions and so on. The existing state-of-the-artmethods are spatial information ignored or spatial scalereduced. To solve these problems, we proposed a novel FullyConvolutional Networks (FCN) model, which is end-to-endlearned on image patches by regressing crowd densitydistribution. Our crowd FCN model can output highprecessioncrowd density map and the crowd quantity can beintegrated by the map. Besides, to handle the problem of sceneperspective distortions, we proposed an unbiased densityground truth generation method. The experiment resultsdemonstrate that our crowd counting method achieved the bestaccuracy on the WorldExpo10 crowd dataset compared withother state-of-the-art methods.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"154 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132462680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
Time Synchronization Method in Quantum Key Distribution System with Automatic Compensation of Polarization Distortions 偏振畸变自动补偿量子密钥分配系统中的时间同步方法
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.68
K. Rumyantsev, Evgeny Rudinskiy
{"title":"Time Synchronization Method in Quantum Key Distribution System with Automatic Compensation of Polarization Distortions","authors":"K. Rumyantsev, Evgeny Rudinskiy","doi":"10.1109/ICMIP.2017.68","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.68","url":null,"abstract":"In this paper, offered a new method of synchronization of the transmitting/receiving and coding stations of autocompensation two-pass quantum key distribution system with phase coding states of photons. This method is based on the use of photonic pulses, which can improve the security level of a synchronization process of quantum key distribution systems. This sync method does not include the division of time frames at the time windows and has a high operating speed. In this paper were synthesized the formulas for calculating the probability characteristics and the parameters of the synchronization process.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116092618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 10
An Improved Single Image Super-Resolution Algorithm Based on Finite Rate of Innovation Theory 基于有限创新率理论的改进单幅图像超分辨算法
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.52
Huashan Ye, Zhizhong Fu, Jin Xu, Keyu Long, Zhongtao Huang
{"title":"An Improved Single Image Super-Resolution Algorithm Based on Finite Rate of Innovation Theory","authors":"Huashan Ye, Zhizhong Fu, Jin Xu, Keyu Long, Zhongtao Huang","doi":"10.1109/ICMIP.2017.52","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.52","url":null,"abstract":"The recently proposed single image super-resolution method which makes use of the theory of sampling signals with finite rate of innovation (FRI) can obtain the state-of-the-art performance. However, the FRI-based method might bring in artifacts in border areas because some values of edge pixels might be overcorrected, leading to extremely sharp edges. To address such issues, we propose an improved FRI-based single image super-resolution method. Specifically, our method can detect the artifacts, and correct the fake value by taking advantage of the correlation of the surrounding pixels. The method would prevent the edges of image from becoming extremely sharp and preserve the true edge texture. The obtained simulation results reveal that the improved FRI-based method can achieve a better performance than the original FRI-method.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123467467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A Log-Polar Feature Guided Iterative Closest Point Method for Image Registration 一种对数极坐标特征引导迭代最近点图像配准方法
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.64
Shihao Zhou, Yun Zhang
{"title":"A Log-Polar Feature Guided Iterative Closest Point Method for Image Registration","authors":"Shihao Zhou, Yun Zhang","doi":"10.1109/ICMIP.2017.64","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.64","url":null,"abstract":"Images can substantially change their appearance and shape when they are acquired using different modalities, with lighting variations or at widely different viewpoints. Even with the state-of-the-art technology, e.g., the generalized dual-bootstrap iterative closest point (GDB-ICP) method, it is still difficult to register those challenging images. To handle this issue, this paper proposes a novel log-polar feature guided iterative closest point (LPF-ICP) method for image registration. In particular, by taking log-polar (LP) image features (including corners and bulbs) as seed, the LPF-ICP method first forms the initial bootstrap regions and the related similar transformation via the matching of LP seeds. Then, driven by the single-scale Harris features (including corners and edges), the proposed method gradually expands the bootstrap regions and updates the transformation estimate until the regions cover the entire image overlap. Finally, the experimental evaluation shows that the LPF-ICP method succeeds in registering all the 22 image pairs contained in the Rensselaer dataset, while the GDB-ICP method only succeeds 19 of them.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"33 11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130588017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Statistical Non-uniformity Correction for Paddlebroom Thermal Infrared Image 桨扫帚热红外图像的统计非均匀性校正
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.2
S. Cao, Yan Li, Cheng Jiang, Guorui Jia
{"title":"Statistical Non-uniformity Correction for Paddlebroom Thermal Infrared Image","authors":"S. Cao, Yan Li, Cheng Jiang, Guorui Jia","doi":"10.1109/ICMIP.2017.2","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.2","url":null,"abstract":"Considering the thermal infrared system and end users need, a statistical non-uniformity correction method is proposed to shorten the time to find suitable and broad ground calibration area. Its basic assumption accords to classically linear model, but it extends traditional operating manner. Specifically and first, for each circle the onboard non-uniform parameter is strictly calculated via sensing two blackbodies in different temperature. Then stack the real paddlebroom data which incorporates atmospheric and motion effect into a virtual one, and fit this small data to reconstruct final coefficients of every pixel statistically. From processing system, the demo cases demonstrate its potential to get a better result (lower variation in real uniform region) without dependency on weather or the classical type of object in image, and largely compress the time to get first image product.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130819872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Visual Pedestrian Tracking from a UAV Platform 基于无人机平台的视觉行人跟踪
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.53
Li Zhang, Zenghui Zhang, Huilin Xiong
{"title":"Visual Pedestrian Tracking from a UAV Platform","authors":"Li Zhang, Zenghui Zhang, Huilin Xiong","doi":"10.1109/ICMIP.2017.53","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.53","url":null,"abstract":"Visual tracking is a challenging problem in computer vision, especially when the camera platform is moving. To track a target pedestrian from a UAV (Unmanned Aerial Vehicle) mobile platform, multiple techniques, such as motion control of UAV, visual detection and tracking are needed. This study presents a new tracking method, which mainly involves object tracking, pedestrian detection and online color feature matching. The algorithm of the proposed method has two main components: a median flow tracker and a pedestrian detection which is based on fast feature pyramids. The proposed tracking method is thoroughly evaluated on various public videos and real scene videos from UAV cameras, and compared with existing state-of-the-art tracking methods to show its effectiveness.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131162758","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
Hybrid Broadcasting Scheme Combining FB+ and DeRe Schemes for Heterogeneous Receivers 基于FB+和DeRe的异构接收机混合广播方案
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.28
Linghao Xiao, Xingjun Wang, Zhanwei Zhong
{"title":"Hybrid Broadcasting Scheme Combining FB+ and DeRe Schemes for Heterogeneous Receivers","authors":"Linghao Xiao, Xingjun Wang, Zhanwei Zhong","doi":"10.1109/ICMIP.2017.28","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.28","url":null,"abstract":"The most efficient way to solve large scale videostreaming problem is periodic broadcasting for popular videos.Early studies mainly focus on reducing client¡¯s waiting timeand buffer demand. Later, researchers focused on the problemof heterogeneity among clients¡¯ reception bandwidths.Solutions can be categorized in two types: ReceptionBandwidth Adaptable (RBA) approach and ReceptionBandwidth Limited (RBL) approach. Both approaches havetheir pros and cons and a recently proposed DeRe scheme hassuccessfully kept the advantages of both approaches together.In this paper, we propose a Hybrid Broadcasting (HyB)scheme which combines FB+ structure and DeRe schemetogether. The proposed scheme not only maintains goodperformance as DeRe scheme in service latency and bufferrequirement, but also overcome the drawbacks of DeRescheme such as large number of video segments to manage andfrequent channel hopping in client¡¯s reception path. HyBscheme is also buffer controllable by adjusting the bufferdemand of its DeRe scheme part.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"32 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120823102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
A High-Capacity Reversible Data Hiding Method in Encrypted Images Based on Block Shifting 一种基于块移位的大容量可逆加密图像数据隐藏方法
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.10
Ruiqi Jiang, Weiming Zhang, Hui Wang, Nenghai Yu
{"title":"A High-Capacity Reversible Data Hiding Method in Encrypted Images Based on Block Shifting","authors":"Ruiqi Jiang, Weiming Zhang, Hui Wang, Nenghai Yu","doi":"10.1109/ICMIP.2017.10","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.10","url":null,"abstract":"Reversible data hiding in encrypted images (RDH-EI) is a new topic drawing more and more attention because of the privacy-preserving requirements from cloud data management. In this paper, a novel reversible data hiding scheme for encrypted image with large embedding capacity is proposed. And it is a client-free RDH-EI scheme in which the RDH algorithm used by the cloud is irrelevant with both the sender and receiver. Different from the previous methods, we propose a block shifting method to transform the original image into the scrambled image which not only looks throughly meaningless, but also obtains a sharp histogram by keeping block correlation meanwhile. For the scrambled image, it is easy for the data hider to reversibly embed data with any plaintext image RDH algorithm such as histogram shifting or lossless compression. The proposed method can achieve real reversibility, that is, data extraction and image recovery are executed with no error. Compared to the previous methods, the experimental results demonstrate that our proposed method increases the embedding capacity in a big range without loss of perfect secrecy.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124048843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
Rapid Forward Vehicle Detection Based on Deformable Part Model 基于可变形零件模型的车辆快速前向检测
2017 2nd International Conference on Multimedia and Image Processing (ICMIP) Pub Date : 2017-03-01 DOI: 10.1109/ICMIP.2017.78
Jing Tang, Zhengkui Lin, Yanan Zhang
{"title":"Rapid Forward Vehicle Detection Based on Deformable Part Model","authors":"Jing Tang, Zhengkui Lin, Yanan Zhang","doi":"10.1109/ICMIP.2017.78","DOIUrl":"https://doi.org/10.1109/ICMIP.2017.78","url":null,"abstract":"This paper first introduce the general method of Deformable Part Model(DPM). In order to improve the speed of front vehicle detection, first, this paper propose a method to accelerate the calculation speed of HOG feature. Second,based on a series of important accelerating mechanisms, this paper mainly use Vector Quantization and K-means clustering algorithm to speed up the HOG feature computing speed, it use Vector Quantization to compress the HOG feature, which is convenient for fast scoring operation. Finally, test the method in different road scenes. Experimental results show that the method can improve the detection speed and meet the real-time requirement without sacrificing the detection precision.","PeriodicalId":227455,"journal":{"name":"2017 2nd International Conference on Multimedia and Image Processing (ICMIP)","volume":"333 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116683914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
首页 上一页 下一页 尾页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信
小红书