In the field of animation and virtualization, there are many existing motion retargeting methods while those methods are not widely applied to practical production. When we transfer a human motion to an avatar in practice, the foot distortion is most obvious. In this paper we present a motion retargeting method for 3D human body with footstep constraints. With the foot end-effector constraints, we solve the footstep slip problem and keep the feet on the ground. To obtain more reasonable results, we also do some smoothing processing with constraints. In this paper we present our experimental results on real captured motion data.
{"title":"A Motion Retargeting Method with Footstep Constraints","authors":"Shaoshuai Xu, Zhixun Su, Xuan Wang","doi":"10.1109/ICDH.2018.00064","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00064","url":null,"abstract":"In the field of animation and virtualization, there are many existing motion retargeting methods while those methods are not widely applied to practical production. When we transfer a human motion to an avatar in practice, the foot distortion is most obvious. In this paper we present a motion retargeting method for 3D human body with footstep constraints. With the foot end-effector constraints, we solve the footstep slip problem and keep the feet on the ground. To obtain more reasonable results, we also do some smoothing processing with constraints. In this paper we present our experimental results on real captured motion data.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125536822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"[Publisher's information]","authors":"","doi":"10.1109/icdh.2018.00066","DOIUrl":"https://doi.org/10.1109/icdh.2018.00066","url":null,"abstract":"","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129058293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Container image correction is an important step in image identification preprocessing for container terminal automation. This paper proposes a method for image correction of multi-angle containers based on computer vision technology. The method includes two steps, the first stage is the filtering and straight-line detection of the container image. And in the second stage, the corresponding points are obtained according to the detected straight-line, and the perspective transformation is performed to achieve the corrective images, which can make it easy to detect and identify image text areas.
{"title":"A Method of Container Image Rectification Based on Computer Vision","authors":"Xiaowei Xu, Jilong Wu, Taofeng Ye, Xiaodong Wang","doi":"10.1109/ICDH.2018.00017","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00017","url":null,"abstract":"Container image correction is an important step in image identification preprocessing for container terminal automation. This paper proposes a method for image correction of multi-angle containers based on computer vision technology. The method includes two steps, the first stage is the filtering and straight-line detection of the container image. And in the second stage, the corresponding points are obtained according to the detected straight-line, and the perspective transformation is performed to achieve the corrective images, which can make it easy to detect and identify image text areas.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123342584","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As an important branch of medical field, Traditional Chinese Medicine(TCM) continues to be explored in data mining research. Taking advantage of machine learning models and deep learning methods, researchers dive into symptom analysis, disease prediction and medicine law. The combination of TCM herbs is the essential basis for compatibility of clinical prescriptions and its research has attracted plenty of attention. However, literature on herb recommendation for clinical diagnosis, to our best knowledge, is slightly lacking. The clinical herbs collocation will be chosen by doctors in consideration of not only the characteristics and pharmacodynamics of the herbs, but also the mutual effects formed with other herbs. Based on the real clinical prescription data, this paper constructs an analytical model to represent the relationship between prescription herbs and syndromes, and develops herb recommendation model. Firstly, by constructing a modeling process based on the LDA topic model, this paper shows the analysis model and presentation method for prescription herbs. Then, based on the mentioned modeling, we propose a doubleend fusion recommendation framework, including methods of adjusting weight proportion and similarity remapping. This research conducts experiments on relevant outpatient medical record data, which confirm that the proposed model can reflect the basic principles of herb combination in clinical diagnosis and the proposed fusion recommendation model has good performance in evaluation metrics.
{"title":"Data-Driven Traditional Chinese Medicine Clinical Herb Modeling and Herb Pair Recommendation","authors":"Gansen Zhao, Xutian Zhuang, Xinming Wang, Weimin Ning, Zijing Li, Jianfei Wang, Qiang Chen, Zefeng Mo, Bingchuan Chen, Huiyan Chen","doi":"10.1109/ICDH.2018.00037","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00037","url":null,"abstract":"As an important branch of medical field, Traditional Chinese Medicine(TCM) continues to be explored in data mining research. Taking advantage of machine learning models and deep learning methods, researchers dive into symptom analysis, disease prediction and medicine law. The combination of TCM herbs is the essential basis for compatibility of clinical prescriptions and its research has attracted plenty of attention. However, literature on herb recommendation for clinical diagnosis, to our best knowledge, is slightly lacking. The clinical herbs collocation will be chosen by doctors in consideration of not only the characteristics and pharmacodynamics of the herbs, but also the mutual effects formed with other herbs. Based on the real clinical prescription data, this paper constructs an analytical model to represent the relationship between prescription herbs and syndromes, and develops herb recommendation model. Firstly, by constructing a modeling process based on the LDA topic model, this paper shows the analysis model and presentation method for prescription herbs. Then, based on the mentioned modeling, we propose a doubleend fusion recommendation framework, including methods of adjusting weight proportion and similarity remapping. This research conducts experiments on relevant outpatient medical record data, which confirm that the proposed model can reflect the basic principles of herb combination in clinical diagnosis and the proposed fusion recommendation model has good performance in evaluation metrics.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134539995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Y. Zhong, Zhixiang Yuan, Shuaijie Zhao, Xiaonan Luo
Based on the traditional Wifi positioning algorithm, the fingerprint database has large redundant information, which is easy to cause dimensionality disaster. In view of this situation, this paper proposes a Wifi positioning method based on ReliefF-SAE. Firstly, the feature selection of the constructed Wifi fingerprint database is carried out, which can remove the redundant information existing in the fingerprint database. Then the reduced fingerprint database is established. The deep neural network is constructed by stacked auto encoder, and pre-training is carried out to obtain a more accurate Wifi indoor positioning model. The positioning experiment is carried out by using the standard database UJIIndoor Loc. The experimental results show that the algorithm can effectively remove the redundant information in the fingerprint database, and at the same time, through the deep neural network learning, better positioning accuracy can be obtained.
{"title":"A Wifi Positioning Method Based on Stack Auto Encoder","authors":"Y. Zhong, Zhixiang Yuan, Shuaijie Zhao, Xiaonan Luo","doi":"10.1109/ICDH.2018.00057","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00057","url":null,"abstract":"Based on the traditional Wifi positioning algorithm, the fingerprint database has large redundant information, which is easy to cause dimensionality disaster. In view of this situation, this paper proposes a Wifi positioning method based on ReliefF-SAE. Firstly, the feature selection of the constructed Wifi fingerprint database is carried out, which can remove the redundant information existing in the fingerprint database. Then the reduced fingerprint database is established. The deep neural network is constructed by stacked auto encoder, and pre-training is carried out to obtain a more accurate Wifi indoor positioning model. The positioning experiment is carried out by using the standard database UJIIndoor Loc. The experimental results show that the algorithm can effectively remove the redundant information in the fingerprint database, and at the same time, through the deep neural network learning, better positioning accuracy can be obtained.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114808649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rain is a common weather that seriously affects the performance of outdoor computer vision applications. The quality of images taken in such weather is very poor. There are several popular methods for the removal of rain streaks from images; one such method is based on physical models and mathematical optimization, and another method is based on deep-learning. However, these methods have their own shortcomings. The optimization-based method is complex, but the result is general. In the deep-learning-based method, some details of the background images are lost through a deep network. In this study, we developed a ResNet and denoising algorithm embedded in the ADMM framework as the background/rain prior. ResNet was trained using synthetic rainy/clear background image pairs as the training data. Then, we divided the images taken in rainy weather into parts with a rainless background and those with the rain streaks. The experiments revealed that the PSNR value of the derain results obtained using a combination of a residual network and the ADMM algorithm was approximately 3% higher than that of the other rain-streak removal algorithms. Moreover, the detailed images obtained were considerably clearer than the details obtained from other rain-streak removal algorithms, and the image quality was better.
{"title":"Domain Knowledge Driven Deep Unrolling for Rain Removal from Single Image","authors":"Ying Ding, Xinwei Xue, Zizhong Wang, Zhiying Jiang, Xin Fan, Zhongxuan Luo","doi":"10.1109/ICDH.2018.00010","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00010","url":null,"abstract":"Rain is a common weather that seriously affects the performance of outdoor computer vision applications. The quality of images taken in such weather is very poor. There are several popular methods for the removal of rain streaks from images; one such method is based on physical models and mathematical optimization, and another method is based on deep-learning. However, these methods have their own shortcomings. The optimization-based method is complex, but the result is general. In the deep-learning-based method, some details of the background images are lost through a deep network. In this study, we developed a ResNet and denoising algorithm embedded in the ADMM framework as the background/rain prior. ResNet was trained using synthetic rainy/clear background image pairs as the training data. Then, we divided the images taken in rainy weather into parts with a rainless background and those with the rain streaks. The experiments revealed that the PSNR value of the derain results obtained using a combination of a residual network and the ADMM algorithm was approximately 3% higher than that of the other rain-streak removal algorithms. Moreover, the detailed images obtained were considerably clearer than the details obtained from other rain-streak removal algorithms, and the image quality was better.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114937380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A residual weighted Chan algorithm is proposed to solve the problem that GPS can not be used for indoor precise positioning. In this paper, the TDOA-base Chan algorithm is derived, and the coordinate position of the measurement point is obtained. Then, the target position is accurately calculated by the residual weight, and the positioning service is provided for the user. The results of laboratory experiments show that the accuracy of the algorithm meets the basic requirements of indoor positioning and improves the error of three-dimensional positioning height.
{"title":"Indoor UWB Location Based on Residual Weighted Chan Algorithm","authors":"Y. Zhong, Ting Wang, Yining Liu, Xiaonan Luo","doi":"10.1109/ICDH.2018.00055","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00055","url":null,"abstract":"A residual weighted Chan algorithm is proposed to solve the problem that GPS can not be used for indoor precise positioning. In this paper, the TDOA-base Chan algorithm is derived, and the coordinate position of the measurement point is obtained. Then, the target position is accurately calculated by the residual weight, and the positioning service is provided for the user. The results of laboratory experiments show that the accuracy of the algorithm meets the basic requirements of indoor positioning and improves the error of three-dimensional positioning height.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129528959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the promotion of the Internet, people can easily retrieve various kinds of education resources on the web. However, current education resources sharing platforms do not support the resources retrieval through the visual information. Therefore, we need to classify the resources which are related to visual characteristics into several categories. In this paper, we propose a novel classification method for resources on Netpad[ http://www.netpad.net.cn/]. We extract the important visual features including graphics features and text features. Then, we use the random forest algorithm to train a valuable model. The results of the experiments indicate that, using graphics features and text features, most of the data are classified correctly, which means that our proposed method can solve the classification problem of Netpad effectively.
{"title":"Online Educational Resources Classification Using Visual Features","authors":"Xiangping Chen, Yancheng Chen, Yonghao Long, Yongsheng Rao, Hao Guan, Mouguang Lin","doi":"10.1109/ICDH.2018.00038","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00038","url":null,"abstract":"With the promotion of the Internet, people can easily retrieve various kinds of education resources on the web. However, current education resources sharing platforms do not support the resources retrieval through the visual information. Therefore, we need to classify the resources which are related to visual characteristics into several categories. In this paper, we propose a novel classification method for resources on Netpad[ http://www.netpad.net.cn/]. We extract the important visual features including graphics features and text features. Then, we use the random forest algorithm to train a valuable model. The results of the experiments indicate that, using graphics features and text features, most of the data are classified correctly, which means that our proposed method can solve the classification problem of Netpad effectively.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132175912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
It is well known that a large number of chaotic and hyperchaotic systems that exhibited research and application prospect in wireless multimedia sensor networks have been discovered successively. This paper proposes a novel noequilibrium hyperchaotic system composed of multi-wing attractors with two positive Lyapunov exponents by means of quadratic function control. It is noteworthy that hidden hyperchaotic attractors can be generated from this noequilibrium system. No-equilibrium analysis of multi-wing hyperchaotic system are also analyzed. Furthermore, by the symmetry conversion with respect to z -axis, intended grid m n -wing hyperchaotic system has been obtained. Finally, electronic circuits of the proposed system are designed for realizing hyperchaotic grid multi-wing attractors, which verifies the feasibility of the theoretical model.
{"title":"A No-Equilibrium Multi-wing Hyperchaotic System with Two Positive Lyapunov Exponents","authors":"Qiang Chen, Chaoxia Zhang, Jinxin Ruan","doi":"10.1109/ICDH.2018.00025","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00025","url":null,"abstract":"It is well known that a large number of chaotic and hyperchaotic systems that exhibited research and application prospect in wireless multimedia sensor networks have been discovered successively. This paper proposes a novel noequilibrium hyperchaotic system composed of multi-wing attractors with two positive Lyapunov exponents by means of quadratic function control. It is noteworthy that hidden hyperchaotic attractors can be generated from this noequilibrium system. No-equilibrium analysis of multi-wing hyperchaotic system are also analyzed. Furthermore, by the symmetry conversion with respect to z -axis, intended grid m n -wing hyperchaotic system has been obtained. Finally, electronic circuits of the proposed system are designed for realizing hyperchaotic grid multi-wing attractors, which verifies the feasibility of the theoretical model.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123626549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
How to satisfy the need of Android users to get high-resolution images quickly becomes a hot issue of research in recent years. We propose a new image interpolation algorithm based on the Loop subdivision. Similar to the refinement of meshes, Subdivision methods can also generate new pixels on images. In order to preserve the sharp edges, We propose a rational subdivision scheme by adjusting the weight coefficients of pixel vertices. This algorithm runs quickly and accurately on the Android platform by using Android NDK.
{"title":"A Subdivision-Based Image Interpolation Method on Android Platform","authors":"Zhihui Yue, Chengming Liu","doi":"10.1109/ICDH.2018.00011","DOIUrl":"https://doi.org/10.1109/ICDH.2018.00011","url":null,"abstract":"How to satisfy the need of Android users to get high-resolution images quickly becomes a hot issue of research in recent years. We propose a new image interpolation algorithm based on the Loop subdivision. Similar to the refinement of meshes, Subdivision methods can also generate new pixels on images. In order to preserve the sharp edges, We propose a rational subdivision scheme by adjusting the weight coefficients of pixel vertices. This algorithm runs quickly and accurately on the Android platform by using Android NDK.","PeriodicalId":117854,"journal":{"name":"2018 7th International Conference on Digital Home (ICDH)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127272231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：