Filters








186 Hits in 1.7 sec

Unified Style Transfer [article]

Guanjie Huang, Hongjian He, Xiang Li, Xingchen Li, Ziang Liu
2021 arXiv   pre-print
Currently, it is hard to compare and evaluate different style transfer algorithms due to chaotic definitions of style and the absence of agreed objective validation methods in the study of style transfer. In this paper, a novel approach, the Unified Style Transfer (UST) model, is proposed. With the introduction of a generative model for internal style representation, UST can transfer images in two approaches, i.e., Domain-based and Image-based, simultaneously. At the same time, a new philosophy
more » ... based on the human sense of art and style distributions for evaluating the transfer model is presented and demonstrated, called Statistical Style Analysis. It provides a new path to validate style transfer models' feasibility by validating the general consistency between internal style representation and art facts. Besides, the translation-invariance of AdaIN features is also discussed.
arXiv:2110.10481v1 fatcat:qsjsmuyvyzfyzb5qkksfxqg5r4

Application of Yolo on Mask Detection Task [article]

Ren Liu, Ziang Ren
2021 arXiv   pre-print
2020 has been a year marked by the COVID-19 pandemic. This event has caused disruptions to many aspects of normal life. An important aspect in reducing the impact of the pandemic is to control its spread. Studies have shown that one effective method in reducing the transmission of COVID-19 is to wear masks. Strict mask-wearing policies have been met with not only public sensation but also practical difficulty. We cannot hope to manually check if everyone on a street is wearing a mask properly.
more » ... xisting technology to help automate mask checking uses deep learning models on real-time surveillance camera footages. The current dominant method to perform real-time mask detection uses Mask-RCNN with ResNet as the backbone. While giving good detection results, this method is computationally intensive and its efficiency in real-time face mask detection is not ideal. Our research proposes a new approach to mask detection by replacing Mask-R-CNN with a more efficient model "YOLO" to increase the processing speed of real-time mask detection and not compromise on accuracy. Besides, given the small volume as well as extreme imbalance of the mask detection datasets, we adopt a latest progress made in few-shot visual classification, simple CNAPs, to improve the classification performance.
arXiv:2102.05402v1 fatcat:q355o5yd25a4tbk4fyl4cbkvhe

Unsupervised Pool-Based Active Learning for Linear Regression [article]

Ziang Liu, Dongrui Wu
2020 arXiv   pre-print
In many real-world machine learning applications, unlabeled data can be easily obtained, but it is very time-consuming and/or expensive to label them. So, it is desirable to be able to select the optimal samples to label, so that a good machine learning model can be trained from a minimum amount of labeled data. Active learning (AL) has been widely used for this purpose. However, most existing AL approaches are supervised: they train an initial model from a small amount of labeled samples,
more » ... new samples based on the model, and then update the model iteratively. Few of them have considered the completely unsupervised AL problem, i.e., starting from zero, how to optimally select the very first few samples to label, without knowing any label information at all. This problem is very challenging, as no label information can be utilized. This paper studies unsupervised pool-based AL for linear regression problems. We propose a novel AL approach that considers simultaneously the informativeness, representativeness, and diversity, three essential criteria in AL. Extensive experiments on 14 datasets from various application domains, using three different linear regression models (ridge regression, LASSO, and linear support vector regression), demonstrated the effectiveness of our proposed approach.
arXiv:2001.05028v1 fatcat:kfxyyeya2vfpbpvbni7w74uiq4

Integrating Informativeness, Representativeness and Diversity in Pool-Based Sequential Active Learning for Regression [article]

Ziang Liu, Dongrui Wu
2020 arXiv   pre-print
In many real-world machine learning applications, unlabeled samples are easy to obtain, but it is expensive and/or time-consuming to label them. Active learning is a common approach for reducing this data labeling effort. It optimally selects the best few samples to label, so that a better machine learning model can be trained from the same number of labeled samples. This paper considers active learning for regression (ALR) problems. Three essential criteria -- informativeness,
more » ... s, and diversity -- have been proposed for ALR. However, very few approaches in the literature have considered all three of them simultaneously. We propose three new ALR approaches, with different strategies for integrating the three criteria. Extensive experiments on 12 datasets in various domains demonstrated their effectiveness.
arXiv:2003.11786v1 fatcat:odhkwfrvubfaldxeemjw2klhqm

Robotic Lime Picking by Considering Leaves as Permeable Obstacles [article]

Heramb Nemlekar, Ziang Liu, Suraj Kothawade, Sherdil Niyaz, Barath Raghavan, Stefanos Nikolaidis
2021 arXiv   pre-print
The problem of robotic lime picking is challenging; lime plants have dense foliage which makes it difficult for a robotic arm to grasp a lime without coming in contact with leaves. Existing approaches either do not consider leaves, or treat them as obstacles and completely avoid them, often resulting in undesirable or infeasible plans. We focus on reaching a lime in the presence of dense foliage by considering the leaves of a plant as 'permeable obstacles' with a collision cost. We then adapt
more » ... e rapidly exploring random tree star (RRT*) algorithm for the problem of fruit harvesting by incorporating the cost of collision with leaves into the path cost. To reduce the time required for finding low-cost paths to goal, we bias the growth of the tree using an artificial potential field (APF). We compare our proposed method with prior work in a 2-D environment and a 6-DOF robot simulation. Our experiments and a real-world demonstration on a robotic lime picking task demonstrate the applicability of our approach.
arXiv:2108.13889v1 fatcat:asdyeuwzqrgyzpfv2vdysv5wt4

Efficient Actuator Failure Avoidance Mobile Charging for Wireless Sensor and Actuator Networks

Jinqi Zhu, Hongrui Yu, Ziang Lin, Nianbo Liu, Huazhi Sun, Ming Liu
2019 IEEE Access  
MING LIU (M'17) received the Ph.D. degree in computer science from Nanjing University, China, in 2006.  ...  ZIANG LIN is currently pursuing the bachelor's degree with the School of Computer and Information Engineering, Tianjin Normal University, China.  ... 
doi:10.1109/access.2019.2931590 fatcat:kp5smqnqmfccnoiamt7ihmddcq

Government Regulations on Closed-Loop Supply Chain with Evolutionarily Stable Strategy

Ziang Liu, Tatsushi Nishi
2019 Sustainability  
The government plays a critical role in the promotion of recycling strategy among supply chain members. The purpose of this study is to investigate the optimal government policies on closed-loop supply chains and how these policies impact the market demand and the returning strategies of manufacturers and retailers. This paper presents a design of closed-loop supply chains under government regulation by considering a novel three-stage game theoretic model. Firstly, Stackelberg models are
more » ... to describe the one-shot game between the manufacturer and the retailer in a local market. Secondly, based on the Stackelberg equilibriums, a repeated and dynamic population game is developed. Thirdly, the government analyzes the population game to find the optimal tax and subsidy policies in the whole market. To solve the proposed model, the idea of backward induction is adopted. The results suggest that, by collecting tax and allocating subsidy, the government can influence the market demands and return rates. The centralized supply chain structure is always preferred for the government and the market. The government prefers to allocate subsidy to low-pollution, low-profit remanufactured products. The environmental attention of the government affects the subsidy policy.
doi:10.3390/su11185030 fatcat:wramun7t35bmfhy6v6sclgs6ba

Multipopulation Ensemble Particle Swarm Optimizer for Engineering Design Problems

Ziang Liu, Tatsushi Nishi, Peter Dabnichki
2020 Mathematical Problems in Engineering  
The MATLAB source codes of MPEPSO are available at https://github.com/zi-ang-liu/MPEPSO.  ... 
doi:10.1155/2020/1450985 fatcat:gb7eo3jvabaapp2ncdqvhgouwy

RainNet: A Large-Scale Dataset for Spatial Precipitation Downscaling [article]

Xuanhong Chen, Kairui Feng, Naiyuan Liu, Yifan Lu, Zhengyan Tong, Bingbing Ni, Ziang Liu, Ning Lin
2020 arXiv   pre-print
Spatial Precipitation Downscaling is one of the most important problems in the geo-science community. However, it still remains an unaddressed issue. Deep learning is a promising potential solution for downscaling. In order to facilitate the research on precipitation downscaling for deep learning, we present the first REAL (non-simulated) Large-Scale Spatial Precipitation Downscaling Dataset, RainNet, which contains 62,424 pairs of low-resolution and high-resolution precipitation maps for 17
more » ... rs. Contrary to simulated data, this real dataset covers various types of real meteorological phenomena (e.g., Hurricane, Squall, etc.), and shows the physical characters - Temporal Misalignment, Temporal Sparse and Fluid Properties - that challenge the downscaling algorithms. In order to fully explore potential downscaling solutions, we propose an implicit physical estimation framework to learn the above characteristics. Eight metrics specifically considering the physical property of the data set are raised, while fourteen models are evaluated on the proposed dataset. Finally, we analyze the effectiveness and feasibility of these models on precipitation downscaling task. The Dataset and Code will be available at https://neuralchen.github.io/RainNet/.
arXiv:2012.09700v2 fatcat:7c3vhna7rjfgpcitfilomhiwki

Measuring and analyzing environment parameters of Dalian Maple Leaf International School

Yuanxiu Wang, Yingxin Zhang, Ziang Lin, Junyi Liu, Z.B. Xu, D.Q. Chen, J.Y. Liu
2019 E3S Web of Conferences  
This paper mainly investigates and analyzes the environment quality of Dalian Maple Leaf International School. Firstly, indoor and outdoor environment parameters are measured, including air quality parameters, noise level parameters, and illuminance parameters. Secondly, by the comparison with environmental quality standards stipulated by China, the environment quality in Dalian Maple Leaf International School is assessed. Finally, some suggestions are given to improve the school environment quality.
doi:10.1051/e3sconf/201913101033 fatcat:ph5cxojwsjdjzpzpxhq4a3paeq

Fabrication of Submicron Structures on Transparent Quartz Glasseswith Improved Optical Properties

Dongyang Zhou, Litong Dong, Ziang Zhang, Mengnan Liu, Ying Wang, Yuegang Fu, Zuobin Wang
2019 Zenodo  
This paper presents a method for the fabrication of submicron structures on transparent quartz glasses to improve optical properties. In this work, the submicron structures were fabricated by two-beam dual exposure laser interference lithography (LIL) and inductively coupled plasma-reactive ion etching (ICP-RIE). The reflectance of less than 5% and the transmittance of more than 95% were achieved in the visible and infrared range of light from 490nm to 1100nm. The experiment results have shown
more » ... hat this method is simple and efficient for the large-area fabrication of submicron structures on transparent quartz glasses with improved optical properties for many applications such as optical components and devices in optical engineering.
doi:10.5281/zenodo.2644017 fatcat:3mcextm3dnbyjjdgkwbkppsjzm

Ground Risk Assessment of UAV Operations Based on Horizontal Distance Estimation under Uncertain Conditions

Yang Liu, Xuejun Zhang, Zhi Wang, Ziang Gao, Chang Liu, Ana C. Teodoro
2021 Mathematical Problems in Engineering  
In this paper a ground safety assessment model is introduced based on the probability estimation of possible impact positions when unmanned aerial vehicle (UAV) crashes on the ground. By incorporating the random uncertainties during the descending process, risks associated with UAV's ground crash are estimated accurately. The number of victims on the ground per flight hour is selected as the indicative index to evaluate the risk levels of the corresponding ground area. We mainly focus on the
more » ... lysis of uncertainties that usually appear in drag coefficient which would generate a great amount of effects on the travelled horizontal distance from the failure point to the impact point on the ground, which further influences the possible impact positions. The drag force in the air, failure velocity of a UAV, and wind effects in the local area are all considered in the proposed model, as well as ground features, including sheltering effects on the ground, UAV parameter settings, and distribution of local population. Uncertainties in drag force when a UAV descends, UAV's initial horizontal and vertical speeds at failure point, and local wind patterns are all considered as the indispensable factors in the proposed model. Especially the probability of fatality once hit by the UAV's debris is explored to make the safety assessment more reliable and valuable. In the end, the actual UAV parameters and official historical weather data are used to estimate the risks in a real operation environment when a failure event happens at a legal flying height. Experimental results are given based on different types of UAVs and random effects in the descent. The results show that all the operations of all kinds of UAVs selected in the validation are so dangerous that the safety of people on the ground cannot be guaranteed, whose value is much bigger than the manned aircraft safety criterion 10−7.
doi:10.1155/2021/3384870 fatcat:qbfn43hv5jgfrdfems6cz5bafe

PREDATOR

Tongping Liu, Chen Tian, Ziang Hu, Emery D. Berger
2014 Proceedings of the 19th ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '14  
Tongping Liu was supported by an internship while at Huawei US Research Center.  ...  Similarly, Liu uses Pin to collect memory access information, and reports total cache miss information [16] . These tools impose about 100 − 200× performance overhead.  ... 
doi:10.1145/2555243.2555244 dblp:conf/ppopp/LiuTHB14 fatcat:63cx5cnk4jfihbrineskcq3b5e

Advances in the toxicology research of microcystins based on Omics approaches

Ya Ma, Haohao Liu, Xingde Du, Ziang Shi, Xiaohui Liu, Rui Wang, Shiyu Zhang, Zhihui Tian, Linjia Shi, Hongxiang Guo, Huizhen Zhang
2021 Environment International  
In female mice, MC-LR could induce pathomorphological changes of the ovary, disturb the estrus cycle and decrease the numbers of primordial follicles (Liu et al., 2021; Wu et al., 2014) .  ...  shown that MC-LR could exert transgenerational toxicity through the placenta and breast milk, resulting in adverse effects on liver, neurodevelopment, immune dysfunction and so forth (Li et al., 2015b; Liu  ... 
doi:10.1016/j.envint.2021.106661 pmid:34077854 fatcat:6auhgi4gwbfmbiqew44huc3g7i

Pool-Based Unsupervised Active Learning for Regression Using Iterative Representativeness-Diversity Maximization (iRDM) [article]

Ziang Liu, Xue Jiang, Hanbin Luo, Weili Fang, Jiajing Liu, Dongrui Wu
2020 arXiv   pre-print
Active learning (AL) selects the most beneficial unlabeled samples to label, and hence a better machine learning model can be trained from the same number of labeled samples. Most existing active learning for regression (ALR) approaches are supervised, which means the sampling process must use some label information, or an existing regression model. This paper considers completely unsupervised ALR, i.e., how to select the samples to label without knowing any true label information. We propose a
more » ... novel unsupervised ALR approach, iterative representativeness-diversity maximization (iRDM), to optimally balance the representativeness and the diversity of the selected samples. Experiments on 12 datasets from various domains demonstrated its effectiveness. Our iRDM can be applied to both linear regression and kernel regression, and it even significantly outperforms supervised ALR when the number of labeled samples is small.
arXiv:2003.07658v2 fatcat:fz23qamr5vejfkiginlj55tx5q
« Previous Showing results 1 — 15 out of 186 results