90 Hits in 1.9 sec

Facial Shape Localization Using Probability Gradient Hints

Zhiheng Niu, Shiguang Shan, Xilin Chen
2009 IEEE Signal Processing Letters  
Niu is with the School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China (e-mail: S. Shan and X.  ... 
doi:10.1109/lsp.2009.2026457 fatcat:ouoecxbrzrgn7j4wtey7t3o44y

Common-specific feature learning for multi-source domain adaptation

Chang Niu, Junyuan Shang, Zhiheng Zhou, Junchu Huang, Tianlei Wang, Xiangwei Li
2020 IET Image Processing  
Multi-source domain adaptation (MDA) aims to leverage knowledge from multiple source domains to improve the classification performance on target domains. Different degrees of distribution discrepancies between every two domains pose a huge challenge to MDA tasks. Most works focus on extracting features shared by all domains, which is critical but not enough to reduce distribution discrepancies. In this paper, we propose a method named as common-specific feature learning (CSFL). Constituting a
more » ... amework of feature learning, CSFL explores a subspace where the combination of common and specific features makes learned representations comprehensive. Based on this framework, we conduct a metric learning method for learning a discriminative feature representation. Considering redundant information caused by source domains is likely to hurt the performance, we impose an effective low-rank constraint to remove the redundant information. Further, we adopt structure consistent constraint to preserve the local structure in each domain. CSFL has obtained about 1-5% improvement of mean accuracy, compared to the state-of-the-art shallow methods. Further, compared with 90.2% and 89.4% of the best baseline deep method, CSFL achieves mean accuracy of 90.8% and 89.7% on the Office-31 and ImageCLEF-DA datasets respectively. The encouraging results validate the effectiveness of our method. IET Image Process.
doi:10.1049/iet-ipr.2019.1712 fatcat:payyipr3v5h5dlkzbdtjhdi3nm


Kechang Niu, Chengjin Chu, Zhiheng Wang
2021 Scientia Sinica Vitae  
Ecology aims to elucidate how organisms interact with other organisms and their environment. As a fundamental concept of ecological research, the term ecological niche was coined to measure the ecological significance of a species in terms of species-specific attributes in relation to the acquisition of environmental resources and impact on the environment. By quantifying species niche, ecologists intended to formulate general rules for controlling various ecological processes, such as species
more » ... istribution, community assembly, dynamics of the food web structure, and functioning. Over the course of a century, the concept of ecological niche has evolved; however, there is limited information about its essence and meaning. Furthermore, we lack knowledge about its quantification in practice, which means that niche-based theories are faced with significant challenges. Hence, we briefly reviewed the evolutionary history of the niche concept and highlighted its fundamental importance in building the theory of community ecology, such as competition exclusion, species coexistence, and community assembly. From the perspective of feedback dynamics between organisms and the environment, we rethink why the ecological niche can be used to explain various existing patterns but predict nearly nothing. We draw a conclusion that ecological niche not only measures the ecological significance of species that drives community dynamics but also presents it as an emerging consequence of the adaptation of species to the community environment. Moreover, it is essentially a conceptual framework, instead of an analytical approach, for understanding ecological rules. By reconciling the perspectives of reductionism and holism, we propose a conceptual framework for a dynamic niche that considers the importance of change in organism attributes, eco-evolutionary feedback, and stochastic processes on the dynamics of species fitness and community assembly. By updating the classical niche concept into a dynamic one, we aim to establish the foundation for the revision of theories of community ecology.
doi:10.1360/ssv-2021-0160 fatcat:tuk27wkpwvcunk4dh3l2dntx7a

Deep Multi-View Concept Learning

Cai Xu, Ziyu Guan, Wei Zhao, Yunfei Niu, Quan Wang, Zhiheng Wang
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
Multi-view data is common in real-world datasets, where different views describe distinct perspectives. To better summarize the consistent and complementary information in multi-view data, researchers have proposed various multi-view representation learning algorithms, typically based on factorization models. However, most previous methods were focused on shallow factorization models which cannot capture the complex hierarchical information. Although a deep multi-view factorization model has
more » ... n proposed recently, it fails to explicitly discern consistent and complementary information in multi-view data and does not consider conceptual labels. In this work we present a semi-supervised deep multi-view factorization method, named Deep Multi-view Concept Learning (DMCL). DMCL performs nonnegative factorization of the data hierarchically, and tries to capture semantic structures and explicitly model consistent and complementary information in multi-view data at the highest abstraction level. We develop a block coordinate descent algorithm for DMCL. Experiments conducted on image and document datasets show that DMCL performs well and outperforms baseline methods.
doi:10.24963/ijcai.2018/402 dblp:conf/ijcai/XuGZNWW18 fatcat:7gwo4xw42bhrhgm4r6vdodob2u

What Shall I Look Like after N Years?

Xiangbo Shu, Jinhui Tang, Luoqi Liu, Zhiheng Niu, Shuicheng Yan
2015 Proceedings of the 23rd ACM international conference on Multimedia - MM '15  
What shall I look like after N years?" In this paper, we present an Auto Age Progression system, which automatically renders a series of aging faces in the future age ranges and generates an aging sequence (aging video) covering the entire life for an individual input. In the offline stage, a set of age-range specific dictionaries are learned from the constructed database, where the dictionary bases corresponding to the same index yet from different dictionaries form a particular aging process
more » ... attern across different age groups, and a linear combination of these patterns expresses a particular personalized aging process. In the online stage, for an input face of an individual, our system renders the aging faces corresponding to different age ranges through the aging dictionaries, and then generates an age progression by the presented face morphing technology.
doi:10.1145/2733373.2807962 dblp:conf/mm/ShuTLNY15 fatcat:kx575vahrfayliuwjqs4vwavtm

Towards Multi-view and Partially-Occluded Face Alignment

Junliang Xing, Zhiheng Niu, Junshi Huang, Weiming Hu, Shuicheng Yan
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
We present a robust model to locate facial landmarks under different views and possibly severe occlusions. To build reliable relationships between face appearance and shape with large view variations, we propose to formulate face alignment as an 1 -induced Stagewise Relational Dictionary (SRD) learning problem. During each training stage, the SRD model learns a relational dictionary to capture consistent relationships between face appearance and shape, which are respectively modeled by the
more » ... indexed image features and the shape displacements for current estimated landmarks. During testing, the SRD model automatically selects a sparse set of the most related shape displacements for the testing face and uses them to refine its shape iteratively. To locate facial landmarks under occlusions, we further propose to learn an occlusion dictionary to model different kinds of partial face occlusions. By deploying the occlusion dictionary into the SRD model, the alignment performance for occluded faces can be further improved. Our algorithm is simple, effective, and easy to implement. Extensive experiments on two benchmark datasets and two newly built datasets have demonstrated its superior performances over the state-of-the-art methods, especially for faces with large view variations and/or occlusions.
doi:10.1109/cvpr.2014.236 dblp:conf/cvpr/XingNHHY14 fatcat:cawc2esijngehirw6ac3aw446e

Context constrained facial landmark localization based on discontinuous Haar-like feature

Xiaowei Zhao, Xiujuan Chai, Zhiheng Niu, Cherkeng Heng, Shiguang Shan
2011 Face and Gesture 2011  
Facial landmark localization is well known as one of the bottlenecks in face recognition. This paper proposes a novel facial landmark localization method, which introduces facial context constrains into cascaded AdaBoost framework. The motivation of our method lies in the basic human physiology observation that not only the local texture information but also the global context information is used together for human to realize the landmark location task. Therefore, in our solution, a novel type
more » ... f Haar-like feature, called discontinuous Haar-like feature, is proposed to characterize the facial context, i.e. the cooccurrence relationship between target facial landmark and other local texture patterns within face region (including other landmarks, facial organs and also smoothing regions). For the locating task, traditional Haar-like features (characterizing local texture information) and discontinuous Haar-like features (characterizing context constrains in global sense) are combined together to form more powerful representations. Through Real AdaBoost learning, distinctive features are selected automatically and used for facial landmark detection. Our experiments on BioID and Cohn-Kanade databases have validated the proposed method by comparing with other state-of-the-art results.
doi:10.1109/fg.2011.5771329 dblp:conf/fgr/ZhaoCNHS11 fatcat:byzchths4jeqpnfyohzjxe7cfi

Enhance ASMs Based on AdaBoost-Based Salient Landmarks Localization and Confidence-Constraint Shape Modeling [chapter]

Zhiheng Niu, Shiguang Shan, Xilin Chen, Bingpeng Ma, Wen Gao
2005 Lecture Notes in Computer Science  
Active Shape Model (ASM) has been recognized as one of the typical methods for image understanding. Simply speaking, it iterates two steps: profile-based landmarks local searching, and statistics-based global shape modeling. We argue that the simple 1D profile matching may not localize landmarks accurately enough, and the unreliable localized landmarks will mislead the following shape matching. Considering these two problems, we propose to enhance ASM from two aspects: (1) in the landmarks
more » ... searching step, we introduce more efficient AdaBoost method to localize some salient landmarks instead of the relatively simple profile matching as in the traditional ASMs; (2) in the global shape modeling step, the confidences of the landmark localization are exploited to constrain the shape modeling and reconstruction procedure by not using those unreliably located landmarks to eliminate their negative effects. We e xperimentally show that the proposed strategies can impressively improve the accuracy of the traditional ASMs.
doi:10.1007/11569947_2 fatcat:eixtwfcssvaknmhvbzyie3elwi

Shape driven kernel adaptation in Convolutional Neural Network for robust facial trait recognition

Shaoxin Li, Junliang Xing, Zhiheng Niu, Shiguang Shan, Shuicheng Yan
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
One key challenge of facial trait recognition is the large non-rigid appearance variations due to some irrelevant real world factors, such as viewpoint and expression changes. In this paper, we explore how the shape information, i.e. facial landmark positions, can be explicitly deployed into the popular Convolutional Neural Network (CNN) architecture to disentangle such irrelevant non-rigid appearance variations. First, instead of using fixed kernels, we propose a kernel adaptation method to
more » ... amically determine the convolutional kernels according to the spatial distribution of facial landmarks, which helps learning more robust features. Second, motivated by the intuition that different local facial regions may demand different adaptation functions, we further propose a tree-structured convolutional architecture to hierarchically fuse multiple local adaptive CNN subnetworks. Comprehensive experiments on WebFace, Morph II and MultiPIE databases well validate the effectiveness of the proposed kernel adaptation method and tree-structured convolutional architecture for facial trait recognition tasks, including identity, age and gender recognition. For all the tasks, the proposed architecture consistently achieves the state-of-the-art performances.
doi:10.1109/cvpr.2015.7298618 dblp:conf/cvpr/LiXNSY15 fatcat:2ui7qsf5xzdjpk7s6ernesqnre

Superselective arterial embolization with drug-loaded microspheres for the treatment of unresectable breast cancer

Zhiheng Wang, Huimin Niu, Zhiyong Li, Jie Zhang, Longjin Sha, Qian Zeng, Xia Liu, Jintang Huang
2019 Gland surgery  
To investigate the short-term efficacy and safety of drug-eluting bead transcatheter arterial chemoembolization (DEB-TACE) using CalliSpheres® microspheres in the treatment of unresectable locally advanced breast cancer (LABC). DEB-TACE using CSM was performed in 15 patients with LABC after failure of medical treatment. The efficacy was evaluated based on the modified Response Evaluation Criteria in Solid Tumors (mRECIST). The postoperative adverse reactions and complications were analyzed. The
more » ... changes of white blood cell (WBC) count, creatine kinase isoenzyme-MB (CK-MB), B-type natriuretic peptide (BNP), and carbohydrate antigen15-3 (CA15-3) before and after treatment were compared by using Wilcoxon signed-rank test. The surgeries were successful in all patients. The subjects were followed up for 2-60 months (median: 10 months). According to the mRECIST, no patient achieved complete remission (CR) 1, 3, and 5 months after surgery, and partial response (PR) was achieved in 9, 11, and 11 cases; also, there were 6, 4, and 2 stable disease (SD) cases, and 0, 0, and 2 progressive disease (PD) cases. The postoperative WBC count, CK-MB level, and BNP level were not significantly different from those before surgery, whereas the CA15-3 level significantly decreased. The main postoperative adverse reactions were pain, fever, and gastrointestinal reactions. No severe adverse reactions were observed. DEB-TACE with CalliSpheres® microspheres is a safe and feasible treatment for LABC. However, more multi-center studies with larger sample sizes are still warranted.
doi:10.21037/gs.2019.12.06 pmid:32042682 pmcid:PMC6989905 fatcat:emaeifgfozgvnbiu7evrf37xsi

Enhancement by Metallic Tube Filling of the Mechanical Properties of Electromagnetic Wave Absorbent Polymethacrylimide Foam

Leilei Yan, Wei Jiang, Chun Zhang, Yunwei Zhang, Zhiheng He, Keyu Zhu, Niu Chen, Wanbo Zhang, Bin Han, Xitao Zheng
2019 Polymers  
By the addition of a carbon-based electromagnetic absorbing agent during the foaming process, a novel electromagnetic absorbent polymethacrylimide (PMI) foam was obtained. The proposed foam exhibits excellent electromagnetic wave-absorbing properties, with absorptivity exceeding 85% at a large frequency range of 4.9–18 GHz. However, its poor mechanical properties would limit its application in load-carrying structures. In the present study, a novel enhancement approach is proposed by inserting
more » ... etallic tubes into pre-perforated holes of PMI foam blocks. The mechanical properties of the tube-enhanced PMI foams were studied experimentally under compressive loading conditions. The elastic modulus, compressive strength, energy absorption per unit volume, and energy absorption per unit mass were increased by 127.9%, 133.8%, 54.2%, and 46.4%, respectively, by the metallic tube filling, and the density increased only by 5.3%. The failure mechanism of the foams was also explored. We found that the weaker interfaces between the foam and the electromagnetic absorbing agent induced crack initiation and subsequent collapses, which destroyed the structural integrity. The excellent mechanical and electromagnetic absorbing properties make the novel structure much more competitive in electromagnetic wave stealth applications, while acting simultaneously as load-carrying structures.
doi:10.3390/polym11020372 pmid:30960356 pmcid:PMC6419214 fatcat:p22clbqw2zfzjf2re2tkk5vdq4

Analyzing Land Use Changes in the Metropolitan Jilin City of Northeastern China Using Remote Sensing and GIS

Dan Hu, Guodong Yang, Qiong Wu, Hongqing Li, Xusheng Liu, Xuefeng Niu, Zhiheng Wang, Qiong Wang
2008 Sensors  
Remote sensing and GIS have been widely employed to study temporal and spatial urban land use changes in southern and southeastern China. However, few studies have been conducted in northeastern regions. This study analyzed land use change and spatial patterns of urban expansion in the metropolitan area of Jilin City, located on the extension of Changbai Mountain, based on aerial photos from 1989 and 2005 Spot images. The results indicated that urban land and transportation land increased
more » ... ically (by 94.04% and 211.20%, respectively); isolated industrial and mining land decreased moderately (by 29.54%); rural residential land increased moderately (by 26.48%); dry land and paddy fields increased slightly (by 15.68% and 11.78%, respectively); forest and orchards decreased slightly (by 5.27% and 4.61%, respectively); grasslands and unused land decreased dramatically (by 99.12% and 86.04%, respectively). Sloped dry land (more than 4 degrees) was mainly distributed on the land below 10 degrees with an east, southeastern and south sunny direction aspect, and most sloped dry land transformed to OPEN ACCESS Sensors 2008, 8 5450 forest was located on an east aspect lower than 12 degrees, while forest changed to dry land were mainly distributed on east and south aspects lower than 10 degrees. A spatial dependency analysis of land use change showed that the increased urban land was a logarithmic function of distance to the Songhua River. This study also provided some data with spatial details about the uneven land development in the upstream areas of Songhua River basin.
doi:10.3390/s8095449 pmid:27873823 pmcid:PMC3705513 fatcat:wc3cxfnvifhgfi6pzwrdgkwzda

Genome-wide CRISPR activation screen identifies novel receptors for SARS-CoV-2 entry [article]

Shiyou Zhu, Ying Liu, Zhuo Zhou, Zhiying Zhang, Xia Xiao, Zhiheng Liu, Ang Chen, Xiaojing Dong, Feng Tian, Shihua Chen, Yiyuan Xu, Chunhui Wang (+7 others)
2021 bioRxiv   pre-print
The ongoing pandemic of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been endangering worldwide public health and economy. SARS-CoV-2 infects a variety of tissues where the known receptor ACE2 is low or almost absent, suggesting the existence of alternative pathways for virus entry. Here, we performed a genome-wide barcoded-CRISPRa screen to identify novel host factors that enable SARS-CoV-2 infection. In addition to known host
more » ... oteins, i.e. ACE2, TMPRSS2, and NRP1, we identified multiple host components, among which LDLRAD3, TMEM30A, and CLEC4G were confirmed as functional receptors for SARS-CoV-2. All these membrane proteins bind directly to spike's N-terminal domain (NTD). Their essential and physiological roles have all been confirmed in either neuron or liver cells. In particular, LDLRAD3 and CLEC4G mediate SARS-CoV-2 entry and infection in a fashion independent of ACE2. The identification of the novel receptors and entry mechanisms could advance our understanding of the multiorgan tropism of SARS-CoV-2, and may shed light on the development of the therapeutic countermeasures against COVID-19.
doi:10.1101/2021.04.08.438924 fatcat:67cll57ymrftnarkta2bulc5wy

Expression of leucine-rich repeat kinase 2 (LRRK2) inhibits the processing of uMtCK to induce cell death in a cell culture model system

Jie Cui, Mei Yu, Jingwen Niu, Zhenyu Yue, Zhiheng Xu
2011 Bioscience Reports  
Mei Yu and Jingwen Niu performed some of the experiments. Zhenyu Yue provided the list of LRRK2 interacting proteins. Zhiheng Xu designed the study and drafted the manuscript.  ...  Expression of leucine-rich repeat kinase 2 (LRRK2) inhibits the processing of uMtCK to induce cell death in a cell culture model system Jie CUI*, Mei YU* †, Jingwen NIU*, Zhenyu YUE † and Zhiheng XU* 1  ... 
doi:10.1042/bsr20100127 pmid:21370995 pmcid:PMC3971885 fatcat:kl2w6toyq5dmlnq7wh5ud7t3sa

MEKK3 coordinates with FBW7 to regulate WDR62 stability and neurogenesis

Dan Xu, Minghui Yao, Yaqing Wang, Ling Yuan, Joerg D. Hoeck, Jingwen Yu, Liang Liu, Yvonne Y. C. Yeap, Weiya Zhang, Feng Zhang, Yinghang Feng, Tiantian Ma (+7 others)
2018 PLoS Biology  
Author Contributions Conceptualization: Zhiheng Xu.  ... 
doi:10.1371/journal.pbio.2006613 pmid:30566428 pmcid:PMC6347294 fatcat:xajfj3mdebhmvn4vu4gkki2ncy
« Previous Showing results 1 — 15 out of 90 results