- 作 者: ( "yuzhuo fu" OR "yu-zhuo fu" )
-
Learning from self-discrepancy via multiple co-teaching for cross-domain person re-identificationMachine Learning (IF 5.414) Pub Date : 2022-06-14 ,DOI:10.1007/s10994-022-06184-xSuncheng Xiang, Yuzhuo Fu, Mengyuan Guan, Ting Liu
Employing clustering strategy to assign unlabeled target images with pseudo labels has become a trend for person re-identification (re-ID) algorithms in domain adaptation. A potential limitation of these clustering-based methods is that they always tend to introduce noisy labels, which will undoubtedly hamper the performance of our re-ID system. To handle this limitation, an intuitive solution is to
...
-
Development and Stiffness Optimization for a Flexible-Tail Robotic FishIEEE Robotics and Automation Letters (IF 4.321) Pub Date : 2021-12-13 ,DOI:10.1109/lra.2021.3134748Ben Lu, Chao Zhou, Jian Wang, Yuzhuo Fu, Long Cheng, Min TanThe integral flexible tail has the potential advantage of lifelike undulating motion. However, due to the complex manufacturing process and difficult modification of structural parameters, its application in robotic fish encounters many challenges. Combining rigid structure and flexible material, this letter proposes a passive flexible fish tail, which can achieve continuous movement and high swimming ...
-
VTBR: Semantic-based Pretraining for Person Re-IdentificationarXiv - CS - Computer Vision and Pattern Recognition Pub Date : 2021-10-11 ,DOI:arxiv-2110.05074Suncheng Xiang, Zirui Zhang, Mengyuan Guan, Hao Chen, Binjie Yan, Ting Liu, Yuzhuo FuPretraining is a dominant paradigm in computer vision. Generally, supervised ImageNet pretraining is commonly used to initialize the backbones of person re-identification (Re-ID) models. However, recent works show a surprising result that ImageNet pretraining has limited impacts on Re-ID system due to the large domain gap between ImageNet and person Re-ID data. To seek an alternative to traditional ...
-
Less is More: Learning from Synthetic Data with Fine-grained Attributes for Person Re-IdentificationarXiv - CS - Computer Vision and Pattern Recognition Pub Date : 2021-09-22 ,DOI:arxiv-2109.10498Suncheng Xiang, Guanjie You, Mengyuan Guan, Hao Chen, Feng Wang, Ting Liu, Yuzhuo FuPerson re-identification (re-ID) plays an important role in applications such as public security and video surveillance. Recently, learning from synthetic data, which benefits from the popularity of synthetic data engine, has attracted attention from both academia and the public eye. However, existing synthetic datasets are limited in quantity, diversity and realisticity, and cannot be efficiently ...
-
Learning from Self-Discrepancy via Multiple Co-teaching for Cross-Domain Person Re-IdentificationarXiv - CS - Computer Vision and Pattern Recognition Pub Date : 2021-04-06 ,DOI:arxiv-2104.02265Suncheng Xiang, Yuzhuo Fu, Mengyuan Guan, Ting LiuEmploying clustering strategy to assign unlabeled target images with pseudo labels has become a trend for person re-identification (re-ID) algorithms in domain adaptation. A potential limitation of these clustering-based methods is that they always tend to introduce noisy labels, which will undoubtedly hamper the performance of our re-ID system. To handle this limitation, an intuitive solution is to ...
-
A cross-layer fault propagation analysis method for edge intelligence systems deployed with DNNsJournal of Systems Architecture (IF 5.836) Pub Date : 2021-02-16 ,DOI:10.1016/j.sysarc.2021.102057Ting Liu, Yuzhuo Fu, Xiaotong Xu, Wei YanTo evaluate the impact of soft errors on convolutional neural networks (CNNs) deployed in edged computation systems, we propose a data-driven assessment strategy to characterize the propagation flow across hardware and software abstraction layers of the system in an interpretable way. Single-bit-flip injections in underlying hardware architecture are performed on virtual embedded system with a CNN-based ...
-
Taking A Closer Look at Synthesis: Fine-grained Attribute Analysis for Person Re-IdentificationarXiv - CS - Computer Vision and Pattern Recognition Pub Date : 2020-10-15 ,DOI:arxiv-2010.08145Suncheng Xiang, Yuzhuo Fu, Guanjie You, Ting LiuPerson re-identification (re-ID) plays an important role in applications such as public security and video surveillance. Recently, learning from synthetic data, which benefits from the popularity of synthetic data engine, has achieved remarkable performance. However, in pursuit of high accuracy, researchers in the academic always focus on training with large-scale datasets at a high cost of time and ...
-
Progressive learning with style transfer for distant domain adaptationIET Image Processing (IF 1.773) Pub Date : 2020-10-14 ,DOI:10.1049/iet-ipr.2020.0166Suncheng Xiang,Yuzhuo Fu,Ting LiuSchool of Microelectronics, Shanghai Jiao Tong UniversityShanghaiPeople's Republic of China
This article studies a novel transfer learning problem termed distant domain transfer learning. Different from traditional transfer learning which assumes there is a close relation between source and target data, in this study, the objective is to execute an unseen and unrelated task based on a labelled data set training previously without any samples from intermediate domains. To this end, the authors
...
-
Multi-level feature learning with attention for person re-identificationMultimedia Tools and Applications (IF 2.577) Pub Date : 2020-08-25 ,DOI:10.1007/s11042-020-09569-zSuncheng Xiang,Yuzhuo Fu,Hao Chen,Wei Ran,Ting LiuPerson re-identification (re-ID) aims to match a specific person in a large gallery with different cameras and locations. Previous part-based methods mainly focus on part-level features with uniform partition, which increases learning ability for discriminative feature but not efficient or robust to scenarios with large variances. To address this problem, in this paper, we propose a novel feature fusion ...
-
Attribute analysis with synthetic dataset for person re-identificationarXiv - CS - Computer Vision and Pattern Recognition Pub Date : 2020-06-12 ,DOI:arxiv-2006.07139Suncheng Xiang, Yuzhuo Fu, Guanjie You, Ting LiuPerson re-identification (re-ID) plays an important role in applications such as public security and video surveillance. Recently, learning from synthetic data, which benefits from the popularity of synthetic data engine, have achieved remarkable performance. However, existing synthetic datasets are in small size and lack of diversity, which hinders the development of person re-ID in real-world scenarios ...
-
Unsupervised person re-identification by hierarchical cluster and domain transferMultimedia Tools and Applications (IF 2.577) Pub Date : 2020-03-30 ,DOI:10.1007/s11042-020-08723-xSuncheng Xiang,Yuzhuo Fu,Mingye Xie,Zefang Yu,Ting LiuPerson re-identification (re-ID) has recently been tremendously boosted due to the advancement of deep convolutional neural networks. Unfortunately, the majority of deep re-ID methods focus on supervised, single-domain re-ID task, while less attention is paid on unsupervised domain adaptation. Therefore, these methods always fail to generalize well to real-world scenarios, which have attracted much ...
-
Spontaneous loss versus stimulation gain in pump-probe microscopy: a proof of concept demonstration.Journal of Biomedical Optics (IF 3.758) Pub Date : 2020-03-01 ,DOI:10.1117/1.jbo.25.3.036501Subir Das,Khalil Ur Rehman,Guan-Yu Zhuo,Fu-Jen Kao
SIGNIFICANCE The large background, narrow dynamic range, and detector saturation have been the common limiting factors in stimulated emission (SE)-based pump-probe microscopy, attributed to the very small signal overriding the very intense laser probe beam. To better differentiate the signal of interest from the background, lock-in detection is used to measure the fluorescence quenching, which is termed
...
-
-
Cross-view gait recognition based on a restrictive triplet networkPattern Recognition Letters (IF 4.757) Pub Date : 2019-07-01 ,DOI:10.1016/j.patrec.2019.04.010Sui-bing Tong,Yu-zhuo Fu,He-fei LingAbstract To overcome the influence of view variations, a restrictive triplet network (RTN) is proposed to solve the problem of cross-view gait recognition in this paper. This network comprises five shared convolutional layers. The restrictive triplet loss is adopted to optimize RTN based on the triplet-based sample groups. These gait samples are selected by a special strategy, so as to make RTN converges ...
-
-
Gait recognition with cross-domain transfer networksJournal of Systems Architecture (IF 5.836) Pub Date : 2019-01-08 ,DOI:10.1016/j.sysarc.2019.01.002Suibing Tong, Yuzhuo Fu, Hefei LingThis paper proposes a novel cross-domain transfer networks (CDTN) that is employed for multi-view gait recognition. CDTN consists of two VGAN layers and a GAN unit. The VGAN layer merges variational autoencoder (VAE) into generative adversarial network (GAN) by replacing the generator of GAN with VAE. Gait energy images (GEIs) are taken as the input of the VGAN layer, and then discriminative loss, ...
-
Multi-View Gait Recognition Based on A Spatial-Temporal Deep Neural NetworkIEEE Access (IF 3.476) Pub Date : 2018-01-01 ,DOI:10.1109/access.2018.2874073Suibing Tong, Yuzhuo Fu, Xinwei Yue, Hefei Ling
This paper proposes a novel spatial–temporal deep neural network (STDNN) that is applied to multi-view gait recognition. The STDNN comprises a temporal feature network (TFN) and a spatial feature network (SFN). In TFN, a feature sub-network is adopted to extract the low-level edge features of gait silhouettes. These features are input to the spatial–temporal gradient (STG) network that adopts a STG
...
-
A Virtual Filter Based Fast Assessment methodology for Fault Tolerant NoCsIEICE Electronics Express Pub Date : 2018-01-01 ,DOI:10.1587/elex.15.20180655Jiajia Jiao, Dezhi Han, Yuzhuo Fu
Network on Chips (NoCs) as the compromising communication infrastructures in many-core system, are suffering the serious Multi-Cell Upsets (MCU) impacts. To accelerate the accurate assessment of the increasing and complex MCU in fault tolerant NoCs, we propose a virtual filter based analytical methodology. The approach converts an original MCU model to a new regenerated MCU model via fault tolerance
...
-
Polarization resolved second harmonic microscopyMethods (IF 4.647) Pub Date : 2017-09-01 ,DOI:10.1016/j.ymeth.2017.06.012Nirmal Mazumder, Gitanjal Deka, Wei-Wen Wu, Ankur Gogoi, Guan-Yu Zhuo, Fu-Jen KaoSecond harmonic (SH) microscopy has proven to be a powerful imaging modality over the past years due to its intrinsic advantages as a multiphoton process with endogenous contrast specificity, which allows pinhole-less optical sectioning, non-invasive observation, deep tissue penetration, and the possibility of easier signal detection at visible wavelengths. Depending on the relative orientation between ...
-
Polarization resolved second harmonic microscopy.Methods (IF 4.647) Pub Date : 2017-06-19 ,DOI:10.1016/j.ymeth.2017.06.012Nirmal Mazumder,Gitanjal Deka,Wei-Wen Wu,Ankur Gogoi,Guan-Yu Zhuo,Fu-Jen KaoSecond harmonic (SH) microscopy has proven to be a powerful imaging modality over the past years due to its intrinsic advantages as a multiphoton process with endogenous contrast specificity, which allows pinhole-less optical sectioning, non-invasive observation, deep tissue penetration, and the possibility of easier signal detection at visible wavelengths. Depending on the relative orientation between ...
-
Three-Dimensional Measurement and Cluster Analysis for Determining the Size Ranges of Chinese Temporomandibular Joint Replacement ProsthesisMedicine (IF 1.817) Pub Date : 2016-02-01 ,DOI:10.1097/md.0000000000002897Lu-Zhu Zhang, Shuai-Shuai Meng, Dong-Mei He, Yu-Zhuo Fu, Ting Liu, Fei-Yu Wang, Min-Jun Dong, Yu-Si ChangFrom the Department of Oral Surgery, College of Stomatology (L-ZZ, D-MH, F-YW, Y-SC); Department of Radiology (M-JD); Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine; and Department of Micro-Nano Electronics, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China (S-SM, Y-ZF, TL).
AbstractThe aim of this study was to investigate the osseous characteristics of Chinese temporomandibular joint (TMJ) and detect the size clusters for total joint prostheses design.Computer tomography (CT) data from 448 Chinese adults (226 male and 222 female, aged from 20 to 83 years, mean age 39.3 years) with 896 normal TMJs were chosen from the Department of Radiology in the Shanghai 9th People's
...
-
A two-level approximate model driven framework for characterizing Multi-Cell Upsets impacts on processorsMicroelectronics Journal (IF 1.992) Pub Date : 2015-12-17 ,DOI:10.1016/j.mejo.2015.11.011Jiajia Jiao, Diana Marculescu, Da-Cheng Juan, Yuzhuo FuSoft error analysis is very significant for a good tradeoff between processor design cost (e.g. area and power) and reliability. In this paper, we propose an approximate model driven framework for efficient soft error analysis in processors. The proposed framework includes: 1) an approximate Probabilistic Graphical Model (PGM) for the Single Bit Upset (SBU) estimation, uses average-and-max policy to ...
-
Exploiting component dependency for accurate and efficient soft error analysis via Probabilistic Graphical ModelsMicroelectronics Reliability (IF 1.418) Pub Date : 2015-01-01 ,DOI:10.1016/j.microrel.2014.09.011Jiajia Jiao, Da-Cheng Juan, Diana Marculescu, Yuzhuo FuAbstract As the technology node continues to scale, soft errors have become a major issue for reliable processor designs. In this paper, we propose a framework that accurately and efficiently estimates the Architectural Vulnerability Factor (AVF) of critical storage structures of a processor. The proposed approach exploits the masking effects between array structure (e.g., register files and Caches) ...
-
A deadlock-free subnetting mechanism for high performance broadcasting in NoCIEICE Electronics Express Pub Date : 2015-01-01 ,DOI:10.1587/elex.12.20150688Xing Han, Yuzhuo Fu, Jiang Jiang, Chang WangSchool of Microelectronics, Shanghai Jiao Tong University
In this letter we present a mechanism to generate subnets for the tasks running on network-on-chip, gaining better performance in broadcast. The present results show that subnetting supplies apparent improvement in the performance of on-chip communications mixed with both unicast and broadcast on manycore processors. However, due to unpredictable task submitting sequence, the applications coming following
...
-
Accelerated assessment of fine-grain AVF in NoC using a Multi-Cell Upsets considered fault injectionMicroelectronics Reliability (IF 1.418) Pub Date : 2014-11-01 ,DOI:10.1016/j.microrel.2014.06.008Jiajia Jiao, Yuzhuo Fu, Shijie WenAbstract With the increasing threat of soft errors induced bits upset, Network on Chip (NoC) as the communication infrastructure in many-core systems has been proven a reliability bottleneck in a fault tolerant parallel system. The often-used metric Architecture Vulnerability Factor (AVF), measures the architecture-level soft error impacts to compromise the design cost of fault tolerant schemes and ...
-
Exploiting and evaluating the potentials of the link addition method for NoC transient error mitigationMicroprocessors and Microsystems (IF 3.503) Pub Date : 2014-02-06 ,DOI:10.1016/j.micpro.2014.01.005Jiajia Jiao, Yuzhuo FuTransient errors in a Network on Chip (NoC) result in some problems such as network blockage, packets loss or incorrect delivery, which would decrease the network throughput and degrade the successful delivery rate. Many fault tolerant mechanisms, such as error correcting code, retransmission and redundancy for the NoC, have been proposed to mitigate transient errors and guarantee the communication ...
-
A refined F-M partitioning algorithm for logic simulationInternational Journal of Simulation and Process Modelling Pub Date : 2012-01-01 ,DOI:10.1504/ijspm.2012.047866Jiafang Wang, Yuzhuo Fu, Jianpei ZhangCircuit partitioning is an efficient way to speed up the parallel simulation and reduce the communication overhead. How to exploit the parallelism inherent in complex and diverse systems is a research hotspot in simulation area. This paper aims to speed-up the simulation of digital and analogue mixed-signal circuits based on parallel and distributed event simulation (PDES) protocols, using the platforms ...
-
Modeling of UHF voltage multiplier for radio‐triggered wake‐up circuitsInternational Journal of Circuit Theory and Applications (IF 2.378) Pub Date : 2010-07-01 ,DOI:10.1002/cta.692Jiang Wang, Liang Dong, Yuzhuo FuWake‐up circuits harvest electromagnetic energy emitted by radio devices, and they can be implemented using ultra‐high frequency (UHF) voltage multipliers. In this letter, we derive an analytical model of the UHF voltage multiplier. The model is compliant with HSPICE circuit simulations, and it allows for short design time of voltage multipliers for radio‐triggered wake‐up circuits. According to this ...
-
Asynchronous complex pipeline design based on ARM instruction setJournal of Shanghai Jiaotong University (Science) Pub Date : 2008-11-05 ,DOI:10.1007/s12204-008-0568-8Bing Wang,Qin Wang,Rui-hua Peng,Yu-zhuo FuThis paper proposes an asynchronous complex pipeline based on ARM-V3 instruction set. Muller pipeline structure is used as prototype, and the factors which may affect pipeline performance are analyzed. To balance the difficulty of asynchronous design and performance analysis, both complete asynchronous and partial asynchronous structures aere designed and compared. Results of comparison with the well-Rnown ...
-
Asynchronous complex pipeline design based on ARM instruction setJournal of Shanghai Jiaotong University (Science) Pub Date : 2008-10-01 ,DOI:10.1007/s12204-008-0568-8Bing Wang, Qin Wang, Rui-hua Peng, Yu-zhuo FuThis paper proposes an asynchronous complex pipeline based on ARM-V3 instruction set. Muller pipeline structure is used as prototype, and the factors which may affect pipeline performance are analyzed. To balance the difficulty of asynchronous design and performance analysis, both complete asynchronous and partial asynchronous structures aere designed and compared. Results of comparison with the well-Rnown ...
高级搜索
我的搜索条件
京公网安备 11010802027423号