In this paper we present a CNN based approach for a real time 3 D-hand pose estimation from the depth sequence.Prior discriminative approaches have achieved remarkable success but are facing two main challenges:Firstl...In this paper we present a CNN based approach for a real time 3 D-hand pose estimation from the depth sequence.Prior discriminative approaches have achieved remarkable success but are facing two main challenges:Firstly,the methods are fully supervised hence require large numbers of annotated training data to extract the dynamic information from a hand representation.Secondly,unreliable hand detectors based on strong assumptions or a weak detector which often fail in several situations like complex environment and multiple hands.In contrast to these methods,this paper presents an approach that can be considered as semi-supervised by performing predictive coding of image sequences of hand poses in order to capture latent features underlying a given image without supervision.The hand is modelled using a novel latent tree dependency model(LDTM)which transforms internal joint location to an explicit representation.Then the modeled hand topology is integrated with the pose estimator using data dependent method to jointly learn latent variables of the posterior pose appearance and the pose configuration respectively.Finally,an unsupervised error term which is a part of the recurrent architecture ensures smooth estimations of the final pose.Experiments on three challenging public datasets,ICVL,MSRA,and NYU demonstrate the significant performance of the proposed method which is comparable or better than state-of-the-art approaches.展开更多
Predictive coding is a promising theoretical framework in neuroscience for understanding information transmission and perception.It posits that the brain perceives the external world through internal models and update...Predictive coding is a promising theoretical framework in neuroscience for understanding information transmission and perception.It posits that the brain perceives the external world through internal models and updates these models under the guidance of prediction errors.Previous studies on predictive coding emphasized top-down feedback interactions in hierarchical multilayered networks but largely ignored lateral recurrent interactions.We perform analytical and numerical investigations in this work on the effects of single-layer lateral interactions.We consider a simple predictive response dynamics and run it on the MNIST dataset of hand-written digits.We find that learning will generally break the interaction symmetry between peer neurons,and that high input correlation between two neurons does not necessarily bring strong direct interactions between them.The optimized network responds to familiar input signals much faster than to novel or random inputs,and it significantly reduces the correlations between the output states of pairs of neurons.展开更多
In this paper, a CMOS image sensor(CIS) is proposed, which can accomplish both decorrelation and entropy coding of image compression directly on the focal plane. The design is based on predictive coding for image deco...In this paper, a CMOS image sensor(CIS) is proposed, which can accomplish both decorrelation and entropy coding of image compression directly on the focal plane. The design is based on predictive coding for image decorrelation. The predictions are performed in analog domain by 2×2 pixel units. Both the prediction residuals and original pixel values are quantized and encoded in parallel. Since the residuals have a peak distribution around zero,the output codewords can be replaced by the valid part of the residuals' binary mode. The compressed bit stream is accessible directly at the output of CIS without extra disposition. Simulation results show that the proposed approach achieves a compression rate of 2. 2 and PSNR of 51 on different test images.展开更多
[Objective] To discuss the effects of major mapping methods for DNA sequence on the accuracy of protein coding regions prediction,and to find out the effective mapping methods.[Method] By taking Approximate Correlatio...[Objective] To discuss the effects of major mapping methods for DNA sequence on the accuracy of protein coding regions prediction,and to find out the effective mapping methods.[Method] By taking Approximate Correlation（AC） as the full measure of the prediction accuracy at nucleotide level,the windowed narrow pass-band filter（WNPBF） based prediction algorithm was applied to study the effects of different mapping methods on prediction accuracy.[Result] In DNA data sets ALLSEQ and HMR195,the Voss and Z-Curve methods are proved to be more effective mapping methods than paired numeric（PN）,Electron-ion Interaction Potential（EIIP） and complex number methods.[Conclusion] This study lays the foundation to verify the effectiveness of new mapping methods by using the predicted AC value,and it is meaningful to reveal DNA structure by using bioinformatics methods.展开更多
Support vector machine （SVM） is powerful to solve some problems such as nonlinear classification, function estimation and density estimation. To consider the chaotic fh （frequency hopping）-code＇s characters in ch...Support vector machine （SVM） is powerful to solve some problems such as nonlinear classification, function estimation and density estimation. To consider the chaotic fh （frequency hopping）-code＇s characters in chaotic dynamic system, the forecasting model of the support vector machine in combination with Takens＇ delay coordinate phase reconstruction of chaotic times is established and the least squares model for large-scale problems is used in local training for this model. Finally, a fh-code series generated by Logistic-Kent mapping is applied to verify the local prediction model. Simulation results show that the high accuracy and fault tolerant SVM model has an excellent performance in predicting the fh code, with a very low mean square error and a high relative coefficient.展开更多
AIM To construct a long non-coding RNA(lnc RNA) signature for predicting hepatocellular carcinoma(HCC) prognosis with high efficiency.METHODS Differentially expressed lnc RNAs(DELs) between HCC specimens and peritumor...AIM To construct a long non-coding RNA(lnc RNA) signature for predicting hepatocellular carcinoma(HCC) prognosis with high efficiency.METHODS Differentially expressed lnc RNAs(DELs) between HCC specimens and peritumor liver specimens were identified using the edge R package to analyze The Cancer Genome Atlas(TCGA) LIHC dataset.Univariate Cox proportional hazards regression was performed to obtain the DELs significantly associated with overall survival(OS) in a training set.These OS-related DELs were further analyzed using a stepwise multivariate Cox regression model.Those lnc RNAs fitted in the multivariate Cox regression model and independently associated with overall survival were chosen to build a prognostic risk formula.The prognostic value ofthis formula was then validated in the test group and the entire cohort and further compared with two previously identified prognostic signatures for HCC.Gene ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses were performed to explore the potential biological functions of the lnc RNAs in the signature.RESULTS Based on lnc RNA expression profiling of 370 HCC patients from the TCGA database,we constructed a 5-lnc RNA signature(AC015908.3,AC091057.3,TMCC1-AS1,DCST1-AS1 and FOXD2-AS1) that was significantly associated with prognosis.HCC patients with high-risk scores based on the expression of the 5 lnc RNAs had significantly shorter survival times compared to patients with low-risk scores in both the training and test groups.Multivariate Cox regression analysis demonstrated that the prognostic value of the 5 lnc RNAs was independent of clinicopathological parameters.A comparison study involving two previously identified prognostic signatures for HCC demonstrated that this 5-lnc RNA signature showed improved prognostic power compared with the other two signatures.Functional enrichment analysis indicated that the 5 lnc RNAs were potentially involved in metabolic processes,fibrinolysis and complement activation.CONCLUSION Our present study constructed a 5-lnc RNA signature that improves survival prediction and can be used as a prognostic biomarker for HCC patients.展开更多
Protein structure prediction is one of the most important problems in structural biology, β-turns are always at the turn of a protein tertiary structure and thus β-turn＇s prediction is a key step in tertiary struct...Protein structure prediction is one of the most important problems in structural biology, β-turns are always at the turn of a protein tertiary structure and thus β-turn＇s prediction is a key step in tertiary structure prediction. There are some methods to predict β-turns based on machine learning techniques such as k-nearest method, neural networks and support vector machine. In this paper, we construct a classifier using double BP networks and put forward two novel methods to code amino acids in the second network. When trained and tested on different datasets, they achieve more accuracy than other coding methods.展开更多
The purpose of software defect prediction is to identify defect-prone code modules to assist software quality assurance teams with the appropriate allocation of resources and labor.In previous software defect predicti...The purpose of software defect prediction is to identify defect-prone code modules to assist software quality assurance teams with the appropriate allocation of resources and labor.In previous software defect prediction studies,transfer learning was effective in solving the problem of inconsistent project data distribution.However,target projects often lack sufficient data,which affects the performance of the transfer learning model.In addition,the presence of uncorrelated features between projects can decrease the prediction accuracy of the transfer learning model.To address these problems,this article propose a software defect prediction method based on stable learning(SDP-SL)that combines code visualization techniques and residual networks.This method first transforms code files into code images using code visualization techniques and then constructs a defect prediction model based on these code images.During the model training process,target project data are not required as prior knowledge.Following the principles of stable learning,this paper dynamically adjusted the weights of source project samples to eliminate dependencies between features,thereby capturing the“invariance mechanism”within the data.This approach explores the genuine relationship between code defect features and labels,thereby enhancing defect prediction performance.To evaluate the performance of SDP-SL,this article conducted comparative experiments on 10 open-source projects in the PROMISE dataset.The experimental results demonstrated that in terms of the F-measure,the proposed SDP-SL method outperformed other within-project defect prediction methods by 2.11%-44.03%.In cross-project defect prediction,the SDP-SL method provided an improvement of 5.89%-25.46% in prediction performance compared to other cross-project defect prediction methods.Therefore,SDP-SL can effectively enhance within-and cross-project defect predictions.展开更多
This paper proposed a back propagation neural network model for predictive block-matching. Predictive block-matching is a way to significantly decrease the computational complexity of motion estimation, but the tradit...This paper proposed a back propagation neural network model for predictive block-matching. Predictive block-matching is a way to significantly decrease the computational complexity of motion estimation, but the traditional prediction model was proposed 26 years ago. It is straight forward but not accurate enough. The proposed back propagation neural network has 5 inputs, 5 neutrons and 1 output. Because of its simplicity, it requires very little calculation power which is negligible compared with existing computation complexity. The test results show 10% - 30% higher prediction accuracy and PSNR improvement up to 0.3 dB. The above advantages make it a feasible replacement of the current model.展开更多
Recently,human motion prediction has gained significant attention and achieved notable success.However,current methods primarily rely on training and testing with ideal datasets,overlooking the impact of variations in...Recently,human motion prediction has gained significant attention and achieved notable success.However,current methods primarily rely on training and testing with ideal datasets,overlooking the impact of variations in the viewing distance and viewing angle,which are commonly encountered in practical scenarios.In this study,we address the issue of model invariance by ensuring robust performance despite variations in view distances and angles.To achieve this,we employed Riemannian geometry methods to constrain the learning process of neural networks,enabling the prediction of invariances using a simple network.Furthermore,this enhances the application of motion prediction in various scenarios.Our framework uses Riemannian geometry to encode motion into a novel motion space to achieve prediction with an invariant viewing distance and angle using a simple network.Specifically,the specified path transport square-root velocity function is proposed to aid in removing the view-angle equivalence class and encode motion sequences into a flattened space.Motion coding by the geometry method linearizes the optimization problem in a non-flattened space and effectively extracts motion information,allowing the proposed method to achieve competitive performance using a simple network.Experimental results on Human 3.6M and CMU MoCap demonstrate that the proposed framework has competitive performance and invariance to the viewing distance and viewing angle.展开更多
高效视频编码(high efficiency video coding,HEVC)相较于上一代编码标准H.264降低了约50%的比特率,但为了提高帧内预测的准确性,HEVC提出的35种预测模式导致计算量大幅增加,对软件和硬件实现均构成了挑战.针对该问题,在HEVC的基础上提...高效视频编码(high efficiency video coding,HEVC)相较于上一代编码标准H.264降低了约50%的比特率,但为了提高帧内预测的准确性,HEVC提出的35种预测模式导致计算量大幅增加,对软件和硬件实现均构成了挑战.针对该问题,在HEVC的基础上提出了一种依据图片纹理方向,结合预测模式之间的关联性来确定帧内预测模式的快速算法.实验结果表明,本算法与HEVC参考软件HM16.20相比,在BD-Rate损失仅为5.79%的情况下,节省46%以上的编码时间,显著降低了帧内预测模式决策的复杂度,便于在嵌入式系统等硬件资源有限的端侧实现算法落地.展开更多
基金supported in part by the Fundamental Research Funds for the Central Universities(WK2350000002)。
文摘In this paper we present a CNN based approach for a real time 3 D-hand pose estimation from the depth sequence.Prior discriminative approaches have achieved remarkable success but are facing two main challenges:Firstly,the methods are fully supervised hence require large numbers of annotated training data to extract the dynamic information from a hand representation.Secondly,unreliable hand detectors based on strong assumptions or a weak detector which often fail in several situations like complex environment and multiple hands.In contrast to these methods,this paper presents an approach that can be considered as semi-supervised by performing predictive coding of image sequences of hand poses in order to capture latent features underlying a given image without supervision.The hand is modelled using a novel latent tree dependency model(LDTM)which transforms internal joint location to an explicit representation.Then the modeled hand topology is integrated with the pose estimator using data dependent method to jointly learn latent variables of the posterior pose appearance and the pose configuration respectively.Finally,an unsupervised error term which is a part of the recurrent architecture ensures smooth estimations of the final pose.Experiments on three challenging public datasets,ICVL,MSRA,and NYU demonstrate the significant performance of the proposed method which is comparable or better than state-of-the-art approaches.
基金supported by the National Natural Science Foundation of China(Grant Nos.11975295 and 12047503)the Chinese Academy of Sciences(Grant Nos.QYZDJ-SSW-SYS018,and XDPD15)
文摘Predictive coding is a promising theoretical framework in neuroscience for understanding information transmission and perception.It posits that the brain perceives the external world through internal models and updates these models under the guidance of prediction errors.Previous studies on predictive coding emphasized top-down feedback interactions in hierarchical multilayered networks but largely ignored lateral recurrent interactions.We perform analytical and numerical investigations in this work on the effects of single-layer lateral interactions.We consider a simple predictive response dynamics and run it on the MNIST dataset of hand-written digits.We find that learning will generally break the interaction symmetry between peer neurons,and that high input correlation between two neurons does not necessarily bring strong direct interactions between them.The optimized network responds to familiar input signals much faster than to novel or random inputs,and it significantly reduces the correlations between the output states of pairs of neurons.
基金Supported by the National Natural Science Foundation of China(No.61036004)Tianjin Research Program of Application Foundation and Advanced Technology(No.13JCQNJC00600)
文摘In this paper, a CMOS image sensor(CIS) is proposed, which can accomplish both decorrelation and entropy coding of image compression directly on the focal plane. The design is based on predictive coding for image decorrelation. The predictions are performed in analog domain by 2×2 pixel units. Both the prediction residuals and original pixel values are quantized and encoded in parallel. Since the residuals have a peak distribution around zero,the output codewords can be replaced by the valid part of the residuals' binary mode. The compressed bit stream is accessible directly at the output of CIS without extra disposition. Simulation results show that the proposed approach achieves a compression rate of 2. 2 and PSNR of 51 on different test images.
基金Supported by Ningxia Natural Science Foundation (NZ1024)the Scientific Research the Project of Ningxia Universities (201027)~~
文摘[Objective] To discuss the effects of major mapping methods for DNA sequence on the accuracy of protein coding regions prediction,and to find out the effective mapping methods.[Method] By taking Approximate Correlation（AC） as the full measure of the prediction accuracy at nucleotide level,the windowed narrow pass-band filter（WNPBF） based prediction algorithm was applied to study the effects of different mapping methods on prediction accuracy.[Result] In DNA data sets ALLSEQ and HMR195,the Voss and Z-Curve methods are proved to be more effective mapping methods than paired numeric（PN）,Electron-ion Interaction Potential（EIIP） and complex number methods.[Conclusion] This study lays the foundation to verify the effectiveness of new mapping methods by using the predicted AC value,and it is meaningful to reveal DNA structure by using bioinformatics methods.
基金the National Natural Science Foundation of China(10577007)Special Fund of Anti-InterferenceTechnology in Tactical Communication Defend Lab(51434020105ZS04).
文摘Support vector machine （SVM） is powerful to solve some problems such as nonlinear classification, function estimation and density estimation. To consider the chaotic fh （frequency hopping）-code＇s characters in chaotic dynamic system, the forecasting model of the support vector machine in combination with Takens＇ delay coordinate phase reconstruction of chaotic times is established and the least squares model for large-scale problems is used in local training for this model. Finally, a fh-code series generated by Logistic-Kent mapping is applied to verify the local prediction model. Simulation results show that the high accuracy and fault tolerant SVM model has an excellent performance in predicting the fh code, with a very low mean square error and a high relative coefficient.
基金Supported by the National Nature Science Foundation of China,No.81702816(to Zhao QJ)Shandong Provincial Natural Science Foundation,No.ZR2017PH030(to Zhao QJ)
文摘AIM To construct a long non-coding RNA(lnc RNA) signature for predicting hepatocellular carcinoma(HCC) prognosis with high efficiency.METHODS Differentially expressed lnc RNAs(DELs) between HCC specimens and peritumor liver specimens were identified using the edge R package to analyze The Cancer Genome Atlas(TCGA) LIHC dataset.Univariate Cox proportional hazards regression was performed to obtain the DELs significantly associated with overall survival(OS) in a training set.These OS-related DELs were further analyzed using a stepwise multivariate Cox regression model.Those lnc RNAs fitted in the multivariate Cox regression model and independently associated with overall survival were chosen to build a prognostic risk formula.The prognostic value ofthis formula was then validated in the test group and the entire cohort and further compared with two previously identified prognostic signatures for HCC.Gene ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses were performed to explore the potential biological functions of the lnc RNAs in the signature.RESULTS Based on lnc RNA expression profiling of 370 HCC patients from the TCGA database,we constructed a 5-lnc RNA signature(AC015908.3,AC091057.3,TMCC1-AS1,DCST1-AS1 and FOXD2-AS1) that was significantly associated with prognosis.HCC patients with high-risk scores based on the expression of the 5 lnc RNAs had significantly shorter survival times compared to patients with low-risk scores in both the training and test groups.Multivariate Cox regression analysis demonstrated that the prognostic value of the 5 lnc RNAs was independent of clinicopathological parameters.A comparison study involving two previously identified prognostic signatures for HCC demonstrated that this 5-lnc RNA signature showed improved prognostic power compared with the other two signatures.Functional enrichment analysis indicated that the 5 lnc RNAs were potentially involved in metabolic processes,fibrinolysis and complement activation.CONCLUSION Our present study constructed a 5-lnc RNA signature that improves survival prediction and can be used as a prognostic biomarker for HCC patients.
基金Supported by the National Natural Science Foundation of China (60773010)
文摘Protein structure prediction is one of the most important problems in structural biology, β-turns are always at the turn of a protein tertiary structure and thus β-turn＇s prediction is a key step in tertiary structure prediction. There are some methods to predict β-turns based on machine learning techniques such as k-nearest method, neural networks and support vector machine. In this paper, we construct a classifier using double BP networks and put forward two novel methods to code amino acids in the second network. When trained and tested on different datasets, they achieve more accuracy than other coding methods.
基金supported by the NationalNatural Science Foundation of China(Grant No.61867004)the Youth Fund of the National Natural Science Foundation of China(Grant No.41801288).
文摘The purpose of software defect prediction is to identify defect-prone code modules to assist software quality assurance teams with the appropriate allocation of resources and labor.In previous software defect prediction studies,transfer learning was effective in solving the problem of inconsistent project data distribution.However,target projects often lack sufficient data,which affects the performance of the transfer learning model.In addition,the presence of uncorrelated features between projects can decrease the prediction accuracy of the transfer learning model.To address these problems,this article propose a software defect prediction method based on stable learning(SDP-SL)that combines code visualization techniques and residual networks.This method first transforms code files into code images using code visualization techniques and then constructs a defect prediction model based on these code images.During the model training process,target project data are not required as prior knowledge.Following the principles of stable learning,this paper dynamically adjusted the weights of source project samples to eliminate dependencies between features,thereby capturing the“invariance mechanism”within the data.This approach explores the genuine relationship between code defect features and labels,thereby enhancing defect prediction performance.To evaluate the performance of SDP-SL,this article conducted comparative experiments on 10 open-source projects in the PROMISE dataset.The experimental results demonstrated that in terms of the F-measure,the proposed SDP-SL method outperformed other within-project defect prediction methods by 2.11%-44.03%.In cross-project defect prediction,the SDP-SL method provided an improvement of 5.89%-25.46% in prediction performance compared to other cross-project defect prediction methods.Therefore,SDP-SL can effectively enhance within-and cross-project defect predictions.
文摘This paper proposed a back propagation neural network model for predictive block-matching. Predictive block-matching is a way to significantly decrease the computational complexity of motion estimation, but the traditional prediction model was proposed 26 years ago. It is straight forward but not accurate enough. The proposed back propagation neural network has 5 inputs, 5 neutrons and 1 output. Because of its simplicity, it requires very little calculation power which is negligible compared with existing computation complexity. The test results show 10% - 30% higher prediction accuracy and PSNR improvement up to 0.3 dB. The above advantages make it a feasible replacement of the current model.
基金supported by the Beijing Municipal Science and Technology Commission and Zhongguancun Science Park Management Committee,No.Z221100002722020National Nature Science Foundation of China,No.62072045Innovation Transfer Fund of Peking University Third Hospital,No.BYSYZHKC2021110。
文摘Recently,human motion prediction has gained significant attention and achieved notable success.However,current methods primarily rely on training and testing with ideal datasets,overlooking the impact of variations in the viewing distance and viewing angle,which are commonly encountered in practical scenarios.In this study,we address the issue of model invariance by ensuring robust performance despite variations in view distances and angles.To achieve this,we employed Riemannian geometry methods to constrain the learning process of neural networks,enabling the prediction of invariances using a simple network.Furthermore,this enhances the application of motion prediction in various scenarios.Our framework uses Riemannian geometry to encode motion into a novel motion space to achieve prediction with an invariant viewing distance and angle using a simple network.Specifically,the specified path transport square-root velocity function is proposed to aid in removing the view-angle equivalence class and encode motion sequences into a flattened space.Motion coding by the geometry method linearizes the optimization problem in a non-flattened space and effectively extracts motion information,allowing the proposed method to achieve competitive performance using a simple network.Experimental results on Human 3.6M and CMU MoCap demonstrate that the proposed framework has competitive performance and invariance to the viewing distance and viewing angle.
文摘高效视频编码(high efficiency video coding,HEVC)相较于上一代编码标准H.264降低了约50%的比特率,但为了提高帧内预测的准确性,HEVC提出的35种预测模式导致计算量大幅增加,对软件和硬件实现均构成了挑战.针对该问题,在HEVC的基础上提出了一种依据图片纹理方向,结合预测模式之间的关联性来确定帧内预测模式的快速算法.实验结果表明,本算法与HEVC参考软件HM16.20相比,在BD-Rate损失仅为5.79%的情况下,节省46%以上的编码时间,显著降低了帧内预测模式决策的复杂度,便于在嵌入式系统等硬件资源有限的端侧实现算法落地.
