Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408775
Aritz Badiola Bengoa, Julen Badiola Martínez, Iñigo Orue Saiz, A. M. Zorrilla
It is well known that each year the amount of elderly people is increasing and the life expectancy is longer in EU countries. So, obviously, it is important to include new tools in the socio-sanitary system to maintain and improve the quality of life of the target group. The objective of the proposal described in this paper is to train cognitive abilities (memory, attention, spacial processing…) to promote independent living though a set of serious games (4). These video games serve as a tool to measure partially the cognitive skills of the users and helps to train and improve them. The tests were carried out with 100 people from 3 countries in different socio-cultural contexts. The first levels of each game are played by users several times with good scores, a fact that does not occur in the higher levels that are usually abandoned. However, the subjective feedback received is very promising as they feel they are having fun.
{"title":"Development and integration of serious games with focus in the training of different cognitive abilities in elderly people to improve their quality of life","authors":"Aritz Badiola Bengoa, Julen Badiola Martínez, Iñigo Orue Saiz, A. M. Zorrilla","doi":"10.1109/ISSPIT51521.2020.9408775","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408775","url":null,"abstract":"It is well known that each year the amount of elderly people is increasing and the life expectancy is longer in EU countries. So, obviously, it is important to include new tools in the socio-sanitary system to maintain and improve the quality of life of the target group. The objective of the proposal described in this paper is to train cognitive abilities (memory, attention, spacial processing…) to promote independent living though a set of serious games (4). These video games serve as a tool to measure partially the cognitive skills of the users and helps to train and improve them. The tests were carried out with 100 people from 3 countries in different socio-cultural contexts. The first levels of each game are played by users several times with good scores, a fact that does not occur in the higher levels that are usually abandoned. However, the subjective feedback received is very promising as they feel they are having fun.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128360840","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408925
Mohammad Alsulami, Raafat S. Elfouly, R. Ammar, Huda Aldosari
Underwater Wireless Sensor Networks(UWSNs) have emerged as a promising technology that is used to monitor underwater environment. Applications of UWSNs are numerous such as oil and gas pipeline monitoring, underwater animal detection, and object of interest detection. Automated Underwater Vehicles (AUVs) have been used to monitor underwater environment [13]. One of the significant challenges of AUVs usage is that it does not meet real-time constraints [15]. Researchers in [1] developed a real-time computing system that can collect, process, and transmit data to a gateway in real-time using a single processing node (computer). Nevertheless, a single computer cannot handle the whole load; Resources and equipment in general are limited. Thus, in this paper, we propose two approaches/algorithms that can group master nodes in the network into groups and allocate a computer for each group. In the first algorithm, we cluster master nodes using bottom-up approach. The process of assigning master nodes, in this approach, to groups is based on the communication range. In the second algorithm, nodes are deployed not only homogeneously but also heterogeneously. We add more constraints in order to make our assumptions are closer to real life. In result section, we provide some insights about our experiments. Simulation results show the merit of our proposed approaches.
{"title":"Deployment of Multiple Computing Systems in Underwater Wireless Sensor Networks","authors":"Mohammad Alsulami, Raafat S. Elfouly, R. Ammar, Huda Aldosari","doi":"10.1109/ISSPIT51521.2020.9408925","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408925","url":null,"abstract":"Underwater Wireless Sensor Networks(UWSNs) have emerged as a promising technology that is used to monitor underwater environment. Applications of UWSNs are numerous such as oil and gas pipeline monitoring, underwater animal detection, and object of interest detection. Automated Underwater Vehicles (AUVs) have been used to monitor underwater environment [13]. One of the significant challenges of AUVs usage is that it does not meet real-time constraints [15]. Researchers in [1] developed a real-time computing system that can collect, process, and transmit data to a gateway in real-time using a single processing node (computer). Nevertheless, a single computer cannot handle the whole load; Resources and equipment in general are limited. Thus, in this paper, we propose two approaches/algorithms that can group master nodes in the network into groups and allocate a computer for each group. In the first algorithm, we cluster master nodes using bottom-up approach. The process of assigning master nodes, in this approach, to groups is based on the communication range. In the second algorithm, nodes are deployed not only homogeneously but also heterogeneously. We add more constraints in order to make our assumptions are closer to real life. In result section, we provide some insights about our experiments. Simulation results show the merit of our proposed approaches.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132950658","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408945
Hanan A. Hassan, O. O. Rabhy, Shimaa A. Mohamed
High-Performance Computing (HPC) system is required for the customers’ needs of high CPU computations. This paper aims to study the communication performance and scalability of ANSYS Fluent applications on HPC system infrastructure. A benchmark of external flow over the aircraft wing is used. The scalability of the ANSYS Fluent application on the HPC system is assessed in terms of core speedup, core rating, and core solver efficiency. The study is conducted on a homogeneous cluster of 5 nodes. Our results reveal that the overall performance of the Intel MPI library is better than OpenMPI and pcmpi libraries for the same experimental design.
{"title":"Performance study of CFD Pressure-based solver on HPC","authors":"Hanan A. Hassan, O. O. Rabhy, Shimaa A. Mohamed","doi":"10.1109/ISSPIT51521.2020.9408945","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408945","url":null,"abstract":"High-Performance Computing (HPC) system is required for the customers’ needs of high CPU computations. This paper aims to study the communication performance and scalability of ANSYS Fluent applications on HPC system infrastructure. A benchmark of external flow over the aircraft wing is used. The scalability of the ANSYS Fluent application on the HPC system is assessed in terms of core speedup, core rating, and core solver efficiency. The study is conducted on a homogeneous cluster of 5 nodes. Our results reveal that the overall performance of the Intel MPI library is better than OpenMPI and pcmpi libraries for the same experimental design.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114612176","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408918
Zijian Zhang, Min Yao
Illumination is an important factor that impairs face recognition. Many algorithms have been proposed to solve the illumination problem. Most algorithms focus on one image information and only use local illumination change, to improve the effects of removing facial illumination. In this paper, we apply the Expected Patch Log Likelihood (EPLL) algorithm to extract illumination weight and we combine it with the Neighboring Radiance Ratio algorithm (NRR) to optimize the initial vector of the Gaussian mixture model, which makes full use of the redundant information in images. The experimental results on the extended Yale B and CMU PIE face databases show that the proposed algorithm can effectively eliminate the influence of illumination on face images and has a high robustness.
{"title":"Illumination Invariant Face Recognition By Expected Patch Log Likelihood","authors":"Zijian Zhang, Min Yao","doi":"10.1109/ISSPIT51521.2020.9408918","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408918","url":null,"abstract":"Illumination is an important factor that impairs face recognition. Many algorithms have been proposed to solve the illumination problem. Most algorithms focus on one image information and only use local illumination change, to improve the effects of removing facial illumination. In this paper, we apply the Expected Patch Log Likelihood (EPLL) algorithm to extract illumination weight and we combine it with the Neighboring Radiance Ratio algorithm (NRR) to optimize the initial vector of the Gaussian mixture model, which makes full use of the redundant information in images. The experimental results on the extended Yale B and CMU PIE face databases show that the proposed algorithm can effectively eliminate the influence of illumination on face images and has a high robustness.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132258636","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408993
Chengdi Cao, Weiqiang Zhang
Cover song identification (CSI) is a challenging task in the music information retrieval (MIR) community. The employment of convolutional neural networks (CNN) have significantly improved the performance of CSI systems, especially CNN designed to be invariant against key transpositions. In this paper, we propose MulKINet, a multi-stage CNN architecture that preserve the property of key invariance while its representational ability is substantially enhanced. Combined with three options for building blocks, channel and temporal attention mechanism, we present an accurate and fast CSI system.
{"title":"MulKINet: Multi-Stage Key-Invariant Convolutional Neural Networks for Accurate and Fast Cover Song Identification","authors":"Chengdi Cao, Weiqiang Zhang","doi":"10.1109/ISSPIT51521.2020.9408993","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408993","url":null,"abstract":"Cover song identification (CSI) is a challenging task in the music information retrieval (MIR) community. The employment of convolutional neural networks (CNN) have significantly improved the performance of CSI systems, especially CNN designed to be invariant against key transpositions. In this paper, we propose MulKINet, a multi-stage CNN architecture that preserve the property of key invariance while its representational ability is substantially enhanced. Combined with three options for building blocks, channel and temporal attention mechanism, we present an accurate and fast CSI system.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124915352","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408862
T. Taniguchi, T. Fujii
This paper proposes power control method in micro operator based small cell system utilizing spectrum database taking into account several scenarios which could happen in future. Here considered is situation where some of user terminals connecting to different base stations are moving, and interference management is carried out taking into account their current location and probability of next moving direction based on two-stage stochastic design for satisfying that the outage SINR of users becomes close to the desired value as possible. Through computer simulations, it has been verified that SINR degradation caused by user movement can be suppressed by the proposed method compared with the fixed location based design avoiding frequent feedback of information on location with less than 10% increase of energy consumption, except the case all users are moving in a system with small number of cells where improvement is relatively small. Some future works like reduction of computational cost are also described.
{"title":"Utilization of Spectrum Database for Power Control in Micro Operator Based Small Cell System by Two-Stage Stochastic Design","authors":"T. Taniguchi, T. Fujii","doi":"10.1109/ISSPIT51521.2020.9408862","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408862","url":null,"abstract":"This paper proposes power control method in micro operator based small cell system utilizing spectrum database taking into account several scenarios which could happen in future. Here considered is situation where some of user terminals connecting to different base stations are moving, and interference management is carried out taking into account their current location and probability of next moving direction based on two-stage stochastic design for satisfying that the outage SINR of users becomes close to the desired value as possible. Through computer simulations, it has been verified that SINR degradation caused by user movement can be suppressed by the proposed method compared with the fixed location based design avoiding frequent feedback of information on location with less than 10% increase of energy consumption, except the case all users are moving in a system with small number of cells where improvement is relatively small. Some future works like reduction of computational cost are also described.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"353 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122764378","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408980
S. Shapoval, B. G. Zapirain, Amaia Méndez Zorilla, Iranzu Mugueta Aguinaga
This paper presents a computer game with automatic body detection and tracking based on the computer camera acquisition for arms and legs rehabilitation. However, with the correct application of different functions and game mechanics, this entertainment can be directed for the good of patients with motor problems, or for elderly people. This research is aimed to study the effectiveness and advantages of computer game platforms in comparison with conventional physical exercise. The experiment involved 13 people. The age of patients ranged from 22-32 (including people with locomotor system problems) to 60-70 years. For each of them the study was divided into two stages: in the first one, the subject was offered to perform a series of physical exercises aimed at warming up the upper and lower body parts. In the second stage, which was held the next day, the subjects performed the same set of exercises, but already in the form of a computer game. After each stage, a pre-prepared survey was conducted which characterized the emotional reaction of each of the subjects to the difference in approaches to rehabilitative physical exercises. The difference in the effectiveness of the game over standard exercises was evaluated by the means of the effectiveness of numerical indicators (table of scores) and the analysis of the therapist. The conducted tests show that at the stage of laboratory testing the average time to pass the current build of the game platform is 512.8 seconds (8.7 minutes), excluding the time to realize the task. The total time including all aspects of the test is 762.6 seconds (12.71 minutes). At the stage of tests with the target audience: the average passing time was 1115.9 seconds (18.5 minutes), the total time was 1809 seconds (30.15 minutes). Evaluating the emotional state and level of motivation shows that player is more interested when hear about an interactive game than the usual physical exercises in front of the mirror. According to the obtained data, it can be concluded that an ordinary set of physical exercises for rehabilitation is more effective if presented as a computer game. The patients have a higher involvement in the process, as well as motivation to improve their results.
{"title":"eSail: a body tracking interactive game for elderly and patients with locomotor system problems","authors":"S. Shapoval, B. G. Zapirain, Amaia Méndez Zorilla, Iranzu Mugueta Aguinaga","doi":"10.1109/ISSPIT51521.2020.9408980","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408980","url":null,"abstract":"This paper presents a computer game with automatic body detection and tracking based on the computer camera acquisition for arms and legs rehabilitation. However, with the correct application of different functions and game mechanics, this entertainment can be directed for the good of patients with motor problems, or for elderly people. This research is aimed to study the effectiveness and advantages of computer game platforms in comparison with conventional physical exercise. The experiment involved 13 people. The age of patients ranged from 22-32 (including people with locomotor system problems) to 60-70 years. For each of them the study was divided into two stages: in the first one, the subject was offered to perform a series of physical exercises aimed at warming up the upper and lower body parts. In the second stage, which was held the next day, the subjects performed the same set of exercises, but already in the form of a computer game. After each stage, a pre-prepared survey was conducted which characterized the emotional reaction of each of the subjects to the difference in approaches to rehabilitative physical exercises. The difference in the effectiveness of the game over standard exercises was evaluated by the means of the effectiveness of numerical indicators (table of scores) and the analysis of the therapist. The conducted tests show that at the stage of laboratory testing the average time to pass the current build of the game platform is 512.8 seconds (8.7 minutes), excluding the time to realize the task. The total time including all aspects of the test is 762.6 seconds (12.71 minutes). At the stage of tests with the target audience: the average passing time was 1115.9 seconds (18.5 minutes), the total time was 1809 seconds (30.15 minutes). Evaluating the emotional state and level of motivation shows that player is more interested when hear about an interactive game than the usual physical exercises in front of the mirror. According to the obtained data, it can be concluded that an ordinary set of physical exercises for rehabilitation is more effective if presented as a computer game. The patients have a higher involvement in the process, as well as motivation to improve their results.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116959875","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408870
Gaël Mahfoudi, F. Morain-Nicolier, F. Retraint, M. Pic
While creating digital image forgeries, it is often necessary to hide an object from the image. For instance, to replace a person on a picture one would first remove the person on the image before inserting the new one. This process of suppression can be performed in many ways but almost always involve recreating some background textures. While recreating it, artists sometimes use common tools and apply smooth transitions to seamlessly blend the result. This operation can decrease the created texture sharpness. In this paper, we explore the possibility to reveal this effect to expose possible forgeries.
{"title":"Object-Removal Forgery Detection Through Reflectance Analysis","authors":"Gaël Mahfoudi, F. Morain-Nicolier, F. Retraint, M. Pic","doi":"10.1109/ISSPIT51521.2020.9408870","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408870","url":null,"abstract":"While creating digital image forgeries, it is often necessary to hide an object from the image. For instance, to replace a person on a picture one would first remove the person on the image before inserting the new one. This process of suppression can be performed in many ways but almost always involve recreating some background textures. While recreating it, artists sometimes use common tools and apply smooth transitions to seamlessly blend the result. This operation can decrease the created texture sharpness. In this paper, we explore the possibility to reveal this effect to expose possible forgeries.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123341323","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408901
Miao Sun, Gurjeet Singh, Patrick Chiang
3D face reconstruction is an attractive topic in computer vision. We have seen dramatic rise in its development recently. Now the state-of-the-art method can reconstruct a face from a single 2D face image freely, which brings a threat to facial security society. Since they are very similar in feature distributions, an efficient work to discriminate reconstructed face and real face is vital. Since Generative Adversarial Nets (GAN) has been proposed by Ian J. Goodfellow in 2014, it is extensively trained to approximate data distributions of many applications. For its adversarial mechanism, GAN shows a powerful generative ability to get the state of art. Inspired by its adversarial mechanism, we propose a similar framework called Anti-GAN to discriminate an adversarial dataset from real 3D face datasets and reconstructed face datasets. Considering the computation of backpropagation, G and D all adopt convolutional neural network architecture. Additionally, experiments show that Anti-GAN is a powerful way to distinguish real faces and reconstructed faces. At the same time, it can also offer robust features for a facial identity task.
三维人脸重建是计算机视觉领域的一个热门课题。最近我们看到它的发展有了惊人的增长。目前最先进的方法可以从单个二维人脸图像中自由地重建人脸，这给人脸安全社会带来了威胁。由于重建人脸和真实人脸在特征分布上非常相似，因此有效区分重建人脸和真实人脸至关重要。自Ian J. Goodfellow于2014年提出生成对抗网络(GAN)以来，它被广泛训练以近似许多应用程序的数据分布。由于其对抗机制，GAN显示出强大的生成能力，达到了最先进的水平。受其对抗机制的启发，我们提出了一个类似的框架Anti-GAN，用于区分真实3D人脸数据集和重建人脸数据集的对抗数据集。考虑到反向传播的计算，G和D都采用卷积神经网络架构。此外，实验表明，Anti-GAN是一种有效区分真实人脸和重建人脸的方法。同时，它还可以为面部识别任务提供强大的功能。
{"title":"Anti-Gan: Discriminating 3D reconstructed and real faces for robust facial Identity in Anti-spoofing Generator Adversarial Network","authors":"Miao Sun, Gurjeet Singh, Patrick Chiang","doi":"10.1109/ISSPIT51521.2020.9408901","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408901","url":null,"abstract":"3D face reconstruction is an attractive topic in computer vision. We have seen dramatic rise in its development recently. Now the state-of-the-art method can reconstruct a face from a single 2D face image freely, which brings a threat to facial security society. Since they are very similar in feature distributions, an efficient work to discriminate reconstructed face and real face is vital. Since Generative Adversarial Nets (GAN) has been proposed by Ian J. Goodfellow in 2014, it is extensively trained to approximate data distributions of many applications. For its adversarial mechanism, GAN shows a powerful generative ability to get the state of art. Inspired by its adversarial mechanism, we propose a similar framework called Anti-GAN to discriminate an adversarial dataset from real 3D face datasets and reconstructed face datasets. Considering the computation of backpropagation, G and D all adopt convolutional neural network architecture. Additionally, experiments show that Anti-GAN is a powerful way to distinguish real faces and reconstructed faces. At the same time, it can also offer robust features for a facial identity task.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130994211","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}
Pub Date : 2020-12-09DOI: 10.1109/ISSPIT51521.2020.9408773
Anuj K. Jain, S. Sahni
Value Iteration (VI) is a powerful, though time consuming, approach to solve reinforcement learning problems modeled as Markov Decision Processes (MDPs). In this paper, we explore strategies to run the sate-of-the-art cache efficient algorithm for VI developed by us [1], [2] on a multicore processor. We demonstrate a speedup of up to 2.59 on a 10-core multiprocessor using 20 threads on popular benchmark data. The speedup for the parallelized portion of the computation is up to 5.89.
{"title":"Value Iteration on Multicore Processors","authors":"Anuj K. Jain, S. Sahni","doi":"10.1109/ISSPIT51521.2020.9408773","DOIUrl":"https://doi.org/10.1109/ISSPIT51521.2020.9408773","url":null,"abstract":"Value Iteration (VI) is a powerful, though time consuming, approach to solve reinforcement learning problems modeled as Markov Decision Processes (MDPs). In this paper, we explore strategies to run the sate-of-the-art cache efficient algorithm for VI developed by us [1], [2] on a multicore processor. We demonstrate a speedup of up to 2.59 on a 10-core multiprocessor using 20 threads on popular benchmark data. The speedup for the parallelized portion of the computation is up to 5.89.","PeriodicalId":111385,"journal":{"name":"2020 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)","volume":"513 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127609888","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：