检索结果-内蒙古大学图书馆

IoU-uniform R-CNN: breaking through the limitations of rpn

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Zhu, Li Xie, Zihao Liu, Liman Tao, Bo Tao, Wenbing National Key Laboratory of Science and Technology on Multi-spectral Information Processing School of Artifcial Intelligence and Automation Huazhong University of Science and Technology Shenzhen Huazhong University of Science and Technology Research Institute Shenzhen518057 China State Key Laboratory of Digital Manufacturing Equipment and Technology School of Mechanical Science and Engineering Huazhong University of Science and Technology WuhanHubei430074 China School of Biomedical Engineering South-Central University for Nationalities Wuhan430074 China

Region Proposal Network (RPN) is the cornerstone of two-stage object detectors, it generates a sparse set of object proposals and alleviates the extrem foreground-background class imbalance problem during training. However, we find that the potential of the detector has not been fully exploited due to the IoU distribution imbalance and inadequate quantity of the training samples generated by RPN. With the increasing intersection over union (IoU), the exponentially smaller numbers of positive samples would lead to the distribution skewed towards lower IoUs, which hinders the optimization of detector at high IoU levels. In this paper, to break through the limitations of RPN, we propose IoU-Uniform R-CNN, a simple but effective method that directly generates training samples with uniform IoU distribution for the regression branch as well as the IoU prediction branch. Besides, we improve the performance of IoU prediction branch by eliminating the feature offsets of RoIs at inference, which helps the NMS procedure by preserving accurately localized bounding box. Extensive experiments on the PASCAL VOC and MS COCO dataset show the effectiveness of our method, as well as its compatibility and adaptivity to many object detection architectures. The code is made publicly availab.e at https://***/zl1994/IoU-Uniform-R-CNN. Copyright © 2019, The Authors. All rights reserved.

关键词： Object detection

Towards real-time eyeblink detection in the wild: Dataset, theory and practices

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Hu, Guilei Xiao, Yang Cao, Zhiguo Meng, Lubin Fang, Zhiwen Zhou, Joey Tianyi Yuan, Junsong National Key Laboratory of Science and Technology on Multi-Spectral Information Processing School of Artificial Intelligence and Automation Huazhong University of Science and Technology China Guangdong Provincial Key Laboratory of Medical Image Processing School of Biomedical Engineering Southern Medical University China School of Energy and Mechanicalelectronic Engineering Hunan University of Humanities Science and Technology China Institute of High Performance Computing A∗STAR Singapore Computer Science and Engineering Department of University at Buffalo State University of New York United States

Effective and real-time eyeblink detection is of widerange applications, such as deception detection, drive fatigue detection, face anti-spoofing. Despite previous efforts, most of existing focus on addressing the eyeblink detection problem under constrained indoor conditions with relative consistent subject and environment setup. Nevertheless, towards practical applications, eyeblink detection in the wild is highly preferred, and of greater challenges. In this paper, we shed the light to this research topic. A lab.lled eyeblink in the wild dataset (i.e., HUST-LEBW) of 673 eyeblink video samples (i.e., 381 positives, and 292 negatives) is first established. These samples are captured from the unconstrained movies, with the dramatic variation on face attribute, head pose, illumination condition, imaging configuration, etc. Then, we formulate eyeblink detection task as a binary spatial-temporal pattern recognition problem. After locating and tracking human eyes using SeetaFace engine and KCF (Kernelized Correlation Filters) tracker respectively, a modified LSTM model able to capture the multi-scale temporal information is proposed to verify eyeblink. A feature extraction approach that reveals the appearance and motion characteristics simultaneously is also proposed. The experiments on HUSTLEBW reveal the superiority and efficiency of our approach. The comparisons with the existing state-of-the-art methods validate the advantages of our manner for eyeblink detection in the wild. Copyright © 2019, The Authors. All rights reserved.

关键词： Long short-term memory

A review of uncertainty quantification in deep learning: Techniques, applications and challenges

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Abdar, Moloud Pourpanah, Farhad Hussain, Sadiq Rezazadegan, Dana Liu, Li Ghavamzadeh, Mohammad Fieguth, Paul Cao, Xiaochun Khosravi, Abbas Rajendra Acharya, U. Makarenkov, Vladimir Nahavandi, Saeid Deakin University Australia College of Mathematics and Statistics Guangdong Key Lab. of Intelligent Information Processing Shenzhen University Shenzhen518060 China Dibrugarh University Dibrugarh India Department of Computer Science and Software Engineering Swinburne University of Technology Melbourne Australia Center for Machine Vision and Signal Analysis University of Oulu Oulu Finland Google research United States Department of Systems Design Engineering University of Waterloo Waterloo Canada State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences Beijing China Department of Electronics and Computer Engineering Ngee Ann Polytechnic Clementi Singapore Department of Computer Science University of Quebec in Montreal MontrealQC Canada

—Uncertainty quantification (UQ) plays a pivotal role in the reduction of uncertainties during both optimization and decision making, applied to solve a variety of real-world applications in science and engineering. Bayesian approximation and ensemble learning techniques are two of the most widely-used UQ methods in the literature. In this regard, researchers have proposed different UQ methods and examined their performance in a variety of applications such as computer vision (e.g., self-driving cars and object detection), image processing (e.g., image restoration), medical image analysis (e.g., medical image classification and segmentation), natural language processing (e.g., text classification, social media texts and recidivism risk-scoring), bioinformatics, etc. This study reviews recent advances in UQ methods used in deep learning, investigates the application of these methods in reinforcement learning, and highlight the fundamental research challenges and directions associated with the UQ field. Copyright © 2020, The Authors. All rights reserved.

关键词： Image segmentation

iSplit LBI: Individualized partial ranking with ties via split LBI

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Xu, Qianqian Sun, Xinwei Yang, Zhiyong Cao, Xiaochun Huang, Qingming Yao, Yuan Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS Microsoft Research Asia State Key Laboratory of Information Security Institute of Information Engineering CAS School of Cyber Security University of Chinese Academy of Sciences School of Computer Science and Tech. University of Chinese Academy of Sciences Key Laboratory of Big Data Mining and Knowledge Management CAS Peng Cheng Laboratory Department of Mathematics Hong Kong University of Science and Technology Hong Kong

Due to the inherent uncertainty of data, the problem of predicting partial ranking from pairwise comparison data with ties has attracted increasing interest in recent years. However, in real-world scenarios, different individuals often hold distinct preferences. It might be misleading to merely look at a global partial ranking while ignoring personal diversity. In this paper, instead of learning a global ranking which is agreed with the consensus, we pursue the tie-aware partial ranking from an individualized perspective. Particularly, we formulate a unified framework which not only can be used for individualized partial ranking prediction, but also be helpful for abnormal user selection. This is realized by a variable splitting-based algorithm called iSplitLBI. Specifically, our algorithm generates a sequence of estimations with a regularization path, where both the hyperparameters and model parameters are updated. At each step of the path, the parameters can be decomposed into three orthogonal parts, namely, abnormal signals, personalized signals and random noise. The abnormal signals can serve the purpose of abnormal user selection, while the abnormal signals and personalized signals together are mainly responsible for individual partial ranking prediction. Extensive experiments on simulated and real-world datasets demonstrate that our new approach significantly outperforms state-of-the-art alternatives. The code is now availiable at https://***/qianqianxu010/ NeurIPS2019-iSplitLBI. Copyright © 2019, The Authors. All rights reserved.

关键词： Forecasting

Fast Semantic Preserving Hashing for Large-Scale Cross-Modal Retrieval

学校读者我要写书评

暂无评论

Fast Semantic Preserving Hashing for Large-Scale Cross-Modal...

IEEE International Conference on Data Mining (ICDM)

作者： Xingzhi Wang Xin Liu Shujuan Peng Yiu-ming Cheung Zhikai Hu Nannan Wang Dept. of Comput. Sci. & Fujian Key Lab. of Big Data Intelligence and Security Huaqiao University Xiamen China State Key Lab. of Integrated Services Networks & School of Telecommun. Eng. Xidian University Xi’an China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou China Dept. of Comput. Sci. and Institute of Research and Continuing Education HK Baptist University Hong Kong SAR China

Most Cross-modal hashing methods do not sufficiently exploit the discrimination power of semantic information when learning hash codes, while often involving time-consuming training procedures for large-scale dataset. To tackle these issues, we first formulate the learning of similarity-preserving hash codes in terms of orthogonally rotating the semantic data to hamming space, and then propose a novel Fast Semantic Preserving Hashing (FSePH) approach to large-scale cross-modal retrieval. Specifically, FSePH introduces an orthonormal basis to regress the targeted hash codes of training examples to their corresponding reasonably relaxed class lab.ls, featuring significantly reducing the quantization error. Meanwhile, an effective optimization algorithm is derived for modality-specific projection function learning and an efficient closed-form solution for hash code learning, which are computationally tractable. Extensive experiments have shown that the proposed FSePH approach runs sufficiently fast, and also significantly improves the retrieval performances over the state-of-the-arts.

关键词：

POSTER: STAR (Space-Time Adaptive and Reductive) Algorithms for Real-World Space-Time Optimality

学校读者我要写书评

暂无评论

ACM SIGPLAN Notices 2017年第8期52卷 455-456页

作者： Tang, Yuan You, Ronghui School of Computer Science School of Software Fudan University Shanghai Key Lab. of Intelligent Information Processing State Key Lab. of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences China

It's important to hit a space-time balance for a real-world algorithm to achieve high performance on modern shared-memory multi-core or many-core systems. However, a large class of dynamic programs with more than $O(1)$ dependency achieve optimality either in space or time, but not both. In the literature, the problem is known as the fundamental space-time tradeoff. By exploiting properly on the runtime system, we show that our STAR (Space-Time Adaptive and Reductive) technique can help these dynamic programs to achieve sublinear parallel time bounds while still maintaining work-, space-, and cache-optimality in a processor- and cache-oblivious fashion. © 2017 ACM.

关键词： Real time systems

Non-reciprocal Cavity Polariton with Atoms Strongly Coupled to Optical Cavity

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Yang, Pengfei Li, Ming Han, Xing He, Hai Li, Gang Zou, Chang-Ling Zhang, Pengfei Qian, Yuhua Zhang, Tiancai State Key Laboratory of Quantum Optics and Quantum Optics Devices Institute of Opto-Electronics Shanxi University Taiyuan030006 China Cas Key Laboratory of Quantum Information University of Science and Technology of China Anhui Hefei230026 China Collaborative Innovation Center of Extreme Optics Shanxi University Taiyuan030006 China Institute of Big Data Science and Industry Shanxi University Taiyuan030006 China Key Lab. of Computational Intelligence and Chinese Information Processing of Ministry of Education Shanxi University Taiyuan030006 China

Breaking the time-reversal symmetry of light is of great importance for fundamental physics and has attracted increasing interest in the study of non-reciprocal photonic devices. Here, we experimentally demonstrate a chiral cavity QED system with multiple atoms strongly coupled to a Fabry-Pérot cavity. By polarizing the internal quantum state of the atoms, the time-reversal symmetry of the atom-cavity interaction is broken. The strongly coupled atom-cavity system can be described by non-reciprocal quasiparticles, i.e., the cavity polariton. When it works in the linear regime, the inherent nonreciprocity makes the system work as a single-photon-level optical isolator. Benefiting from the collective enhancement of multiple atoms, an isolation ratio exceeding 30 dB on the single-quanta level (∼ 0.1 photon on average) is achieved. The validity of the non-reciprocal device under zero magnetic field and the reconfigurability of the isolation direction are also experimentally demonstrated. Moreover, when the cavity polariton works in the nonlinear regime, the quantum interference between polaritons with weak anharmonicity induces non-reciprocal nonclassical statistics of cavity transmission from coherent probe light. Copyright © 2019, The Authors. All rights reserved.

关键词： Atoms

A Fast Proportionate RLS Adaptive Equalization for Underwater Acoustic Communications

学校读者我要写书评

暂无评论

A Fast Proportionate RLS Adaptive Equalization for Underwate...

OCEANS - Europe

作者： Zhen Qin Jun Tao Feng Tong Hongtao Zhang Fengzhong Qu Xiao Han Key Laboratory of Underwater Acoustic Signal Processing of the Ministry of Education School of Information Science and Engineering Southeast University Nanjing China State Key Lab. of Acoustics Institute of Acoustics Chinese Academy of Sciences Beijing China College of Ocean and Earth Sciences Xiamen University Xiamen China Hangzhou Appiled Acoustics Research Institute Hangzhou China Key Laboratory of Ocean Observation-Imaging Testbed of Zhejiang Province Zhejiang University Zhoushan China Acoustic Science and Technology Laboratory Harbin Engineering University Harbin China

In this paper, a low-complexity recursive least squares (RLS) type sparse direct adaptive equalizer (DAE) is proposed for multiple-input multiple-output (MIMO) underwater acoustic (UWA) communications. The underlying proportionate RLS (PRLS) adaptive filter algorithm is motivated by the idea of proportionate updating (PU) originated in the improved proportionate normalized least mean squares (IPNLMS) adaptive filter scheme. To overcome the high complexity of the PRLS that is quadratic in the filter size, a fast implementation is developed in a way similar to the design of the stable fast transversal filter (SFTF) as a low-complexity approximation of the standard RLS adaptive filter algorithm. The resulting fast version of the PRLS is named the proportionate SFTF (PSFTF). The PSFTF is then adopted to update coefficients of a linear equalizer (LE), which was tested by experimental data collected in at-sea UWA communication trials. Experimental results showed the PSFTFDAE achieves faster convergence and better performance than existing SFTF-DAE and RLS-DAE.

关键词： Complexity theory MIMO communication Standards Equalizers Channel estimation Underwater acoustics Signal processing algorithms

Research on Adaptation Criteria Generation based on Large Data Mining 7

学校读者我要写书评

暂无评论

Research on Adaptation Criteria Generation based on Large Da...

2016 IEEE Chinese Guidance, Navigation and Control Conference (IEEE CGNCC2016)

作者： Yun Hong-quan XU Li DOU Hao MING De-lie National Key Laboratory of Science and Technology on Aerospace Intelligence Control Beijing Aerospace Automatic Control Institute State Key Laboratory for Multi-spectral Information Processing Technologies School of AutomationHuazhong University of Science and Technology

ISBN: (纸本)9781467383196

This paper performs research on adaptation analysis based on large amount of multi-source remote sensing *** view of different demands from different task background,the research is firstly focused on how to analyze the data in computer *** achieve this,the feature parameters of target areas are extracted from different target area geographic *** combination of ORACLE database engine,data mining technology is used to carry out the target area adaptation assessment,and extract corresponding adaptation *** test the trained adaptation criteria on multi-source geographic information data of different target *** results show that the resulting criterion has certain coincidence rate and robustness.

关键词： image Research on Adaptation Criteria Generation based on Large Data Mining