发布求助
文献互助 智能选刊 最新文献

2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)最新文献

筛选
英文 中文
Artifacts reduction for very low bitrate image compression with generative adversarial networks 用生成对抗网络减少非常低比特率图像压缩的伪影
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966158
Sébastien Hamis, T. Zaharia, Olivier Rousseau
{"title":"Artifacts reduction for very low bitrate image compression with generative adversarial networks","authors":"Sébastien Hamis, T. Zaharia, Olivier Rousseau","doi":"10.1109/ICCE-Berlin47944.2019.8966158","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966158","url":null,"abstract":"Image compression at very low bitrate has known a new dawn since the democratization of convolutional neural networks (CNNs). A promising approach consists in using an « end-to-end » neural network-based compression scheme, performing both coding and decoding phases. Such techniques lead to perceptually convincing results, notably with GAN-based architectures. However, due to the nature of GAN, alongside being computationally expensive on the encoder side, such schemes cannot ensure the preservation of fine details, such as small digits or letters on an ID card. To overcome this limitation, we propose a new model, specifically trained to reduce artifacts resulting from strong lossy compressions. An advantage of the proposed approach comes from the fact that it can be interpreted as a post-processing step that can easily be added to any compression scheme without modifying the codec. Experimental results show that our model perceptually outperforms the state-of-the-art compression standards for very low bitrates.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131747932","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}
引用次数: 1
Multiunit automotive perception framework: Synergy between AI and deterministic processing 多单元汽车感知框架:人工智能与确定性处理之间的协同作用
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966168
Nives Kaprocki, Gordana S. Velikic, N. Teslic, Momcilo Krunic
{"title":"Multiunit automotive perception framework: Synergy between AI and deterministic processing","authors":"Nives Kaprocki, Gordana S. Velikic, N. Teslic, Momcilo Krunic","doi":"10.1109/ICCE-Berlin47944.2019.8966168","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966168","url":null,"abstract":"Since neural networks were first introduced into automotive systems, safety has been a major concern. The prevailing safety standard in the automotive industry, ISO26262, does not fully define testing and verification methods for software based on deep learning. In this paper, we propose a multiunit perception framework that increases the determinism of automotive systems incorporating deep learning. Our approach relies on ASIL decomposition and algorithm diversification, which are enabled through the utilization of multiple low ASIL perception units and one high ASIL monitor unit. In addition to the framework concept, we specify how each component can be mapped to appropriate hardware and software platforms. The practical feasibility of the perception framework is demonstrated with a proof of concept prototype.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"166 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116306267","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}
引用次数: 2
Fast and Light-weight Binarized Neural Network Implemented in an FPGA using LUT-based Signal Processing and its Time-domain Extension for Multi-bit Processing 基于lut的信号处理及其多比特处理的时域扩展在FPGA上实现了快速轻量级二值化神经网络
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966187
Ryuji Fuchikami, F. Issiki
{"title":"Fast and Light-weight Binarized Neural Network Implemented in an FPGA using LUT-based Signal Processing and its Time-domain Extension for Multi-bit Processing","authors":"Ryuji Fuchikami, F. Issiki","doi":"10.1109/ICCE-Berlin47944.2019.8966187","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966187","url":null,"abstract":"Fast and small-resourced implementation of convolutional neural network (CNN) into a field-programmable gate array (FPGA) was realized using a binarized neural network (NN). We propose a set of neuron and network models optimized for fully binarized implementation of general NNs using the look-up-tables (LUTs) in modern FPGAs, which is herein referred to as sparse-LUT model. Arrayed MNIST data images of more than 40 characters input from a camera were recognized with 92.8% accuracy and classified with the colored marks on the characters on organic light-emitting diode (OLED) display images with 1-ms cycle time, <1.0-ms delay in the LUT-based CNN recognition, and <2-ms total time delay. In combination with stochastic time-divided signal processing, binarized signals in this model can be extended for processing multi-bit (analogue-like) signals in an oversampling manner with increased recognition accuracies up to 98.6% in MNIST and 58.3% in CIFAR-10 image data sets using CNN. The source codes for the binarized NN core were released and open-sourced.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116338698","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}
引用次数: 6
Using Class-Dependent Post-Processing to Improve AI for Automotive Front View Camera 使用类相关后处理改进汽车前视摄像头的AI
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966233
Milorad Marković, M. Milosevic, Ninoslav Jovanov
{"title":"Using Class-Dependent Post-Processing to Improve AI for Automotive Front View Camera","authors":"Milorad Marković, M. Milosevic, Ninoslav Jovanov","doi":"10.1109/ICCE-Berlin47944.2019.8966233","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966233","url":null,"abstract":"Deep learning approaches in artificial intelligence are being increasingly applied to many modern applications, including autonomous driving. Semantic segmentation is one of the promising ways to map vehicle environment utilizing a front view camera. In this paper we propose a post-processing class-dependent method to improve accuracy and configurability of semantic segmentation, arguing that this method is appropriate for front view application. We utilized morphological transformations to overcome the effects of noise and class overlap as well as to improve visibility of classes which had limited presence in the input dataset. Using this method, the output can increase in accuracy to up to 3.88%.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128909024","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}
引用次数: 0
Stage UI: Live TV Application Tailored for Mobile Devices 舞台UI:为移动设备量身定制的电视直播应用
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966166
Lena Bobic, Natasa Vukota, Gordana S. Velikic, I. Kastelan
{"title":"Stage UI: Live TV Application Tailored for Mobile Devices","authors":"Lena Bobic, Natasa Vukota, Gordana S. Velikic, I. Kastelan","doi":"10.1109/ICCE-Berlin47944.2019.8966166","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966166","url":null,"abstract":"Change in cultural trends of multimedia deployment has put a new weight to rolls and uses of mobile devices. The size of the screen and touch possibilities provide additional dimensions in designs of the user interfaces compared to the TV exclusive applications. In this paper we have explored new possibilities to extend the STAGE TV application in two directions: menus layout and deployed devices. This approach we named Stage UI - an interface for hybrid TV receivers based on the Android operating system for TV. This interface has modern TV functionalities that allow users to interact with Google services, and to promote and prioritize content. This solution is available for both, over-the-top devices, and for hybrid solutions via Android TV Input Frameworks. Stage UI is completely modular and customizable to the operator as the end user. Due to its interface scalability, it is also adaptable to mobile devices. In this paper we present one solution how to adapt the Live TV application for mobile devices.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130755945","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}
引用次数: 0
A Machine-Learning Based Emotion Recognition System in Patients with Parkinson’s Disease 基于机器学习的帕金森病患者情绪识别系统
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966224
M. Capecci, L. Ciabattoni, G. Foresi, A. Monteriù, L. Pepa
{"title":"A Machine-Learning Based Emotion Recognition System in Patients with Parkinson’s Disease","authors":"M. Capecci, L. Ciabattoni, G. Foresi, A. Monteriù, L. Pepa","doi":"10.1109/ICCE-Berlin47944.2019.8966224","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966224","url":null,"abstract":"In this paper, a Machine-Learning Based Emotion Recognition System in patients with Parkinson’s disease is presented. The development of this system is composed of three steps. Firstly, each user is required to execute an experimental protocol while a simple device (i.e., smartwatch), worn on the wrist, collects data. During the experimental protocol, a nine-point clinical scale and a commercial emotion recognition software have been used to identify emotions. Secondly, from smartwatch data, features extraction is implemented. Lastly, a Machine Learning Algorithm (MLA) is trained with extracted features as input and emotion classes as output.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123191117","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}
引用次数: 3
Double Threshold Soft Decision Rule for Improving Performance of Cognitive Radio Networks 提高认知无线网络性能的双阈值软决策规则
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966232
Hikmat N. Abdullah, Nuha Sh. Bakr, A. Valenzuela
{"title":"Double Threshold Soft Decision Rule for Improving Performance of Cognitive Radio Networks","authors":"Hikmat N. Abdullah, Nuha Sh. Bakr, A. Valenzuela","doi":"10.1109/ICCE-Berlin47944.2019.8966232","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966232","url":null,"abstract":"We propose an efficient soft decision rule to enhance the detection of unused spectra and to improve cognitive radio (CR) throughput. The performance of this rule is evaluated for additive white Gaussian noise and Rayleigh fading channels and compared with the square law and maximal ratio combining rules. For a signal to noise ratio (SNR) of 2dB, the proposed rule enhances the probability of a false alarm by 68% and 95% respectively in case of additive white Gaussian noise and by 53% and 92% in case of Rayleigh fading channels. The simulation results also show that the proposed method enhances throughput compared to the other rules. Over Rayleigh fading channels, throughput is enhanced by 210% compared to square law combining rule and 40% compared to the maximal ratio combining rule.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125222609","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}
引用次数: 3
Estimating Power Consumption of Air-conditioners Using a Sequence-to-sequence Model 基于序列对序列模型的空调功耗估算
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966143
Inhwan Hwang, H. Cho, Yunhu Ji, Huijung Kim
{"title":"Estimating Power Consumption of Air-conditioners Using a Sequence-to-sequence Model","authors":"Inhwan Hwang, H. Cho, Yunhu Ji, Huijung Kim","doi":"10.1109/ICCE-Berlin47944.2019.8966143","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966143","url":null,"abstract":"We suggest a neural network based power estimation algorithm for air-conditioners exploiting a sequence-to-sequence model which is widely used in the field of natural language processing. Our proposed method exploits both past usage patterns and weather forecasts for better prediction because air-conditioners have high dependency on the environmental factors. Our proposed algorithm is trained and validated on real-world data collected in various regions of the Republic of Korea. For validation, we adopt two evaluation metrics, mean absolute error and mean absolute percentage error. We accordingly suggest two different types of accuracy metrics practical for end users. Applying the proposed method to real-world data from air-conditioners, we have 5.7 kW of mean absolute error and 23 percent of mean absolute percentage error. Accuracy out of each metric is 85.02 percent and 74.65 percent respectively which outperforms the conventional regression algorithms.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127772040","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}
引用次数: 1
Light field image compression using Versatile Video Coding 光场图像压缩使用多功能视频编码
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966199
Vasileios Avramelos, J. D. Praeter, G. Wallendael, P. Lambert
{"title":"Light field image compression using Versatile Video Coding","authors":"Vasileios Avramelos, J. D. Praeter, G. Wallendael, P. Lambert","doi":"10.1109/ICCE-Berlin47944.2019.8966199","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966199","url":null,"abstract":"Light field cameras (or plenoptic cameras) made it possible to not only capture information about the light intensity in a scene, but also the light direction traveling in space. The resulting light field images added two extra spatial dimensions to conventional 2D images by capturing a scene from different angles. Light field technology is expected to deliver the needed six-degrees-of-freedom in virtual reality, and it is a cornerstone technology in holography. Due to the vast amount of data a light field needs to capture for representing high-resolution images, the need of new coding solutions has been considered. While coding standards aiming at plenoptic images are something to be expected in the near future, video coding standards are one of the most efficient solutions to compress light fields. There are countless scanning topologies for restructuring a light field image into a 2D video, and similarly, numerous prediction structures to be used for inter-prediction. In this paper, we provide a comparison of light field image coding performance using the three latest generations of video compression standards. Moreover, we assess a set of coding structures aiming for optimal compression rates as well as random access capabilities. Results show the necessary trade-off between bandwidth and potential random access between light field views at decoding level. Bitrate savings up to 25% were achieved at certain cases where the proposed prediction structure is used over a common scheme used in the literature. The findings of this overview provide a useful benchmark for future development of light field image coding solutions using video coding standards.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134089806","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}
引用次数: 3
A Route Recommendation Method Based on Personal Preferences by Monte-Carlo Tree Search 基于蒙特卡罗树搜索的个人偏好路线推荐方法
2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin) Pub Date : 2019-09-01 DOI: 10.1109/ICCE-Berlin47944.2019.8966146
Yuta Ishizaki, Toshinori Takayama, N. Togawa
{"title":"A Route Recommendation Method Based on Personal Preferences by Monte-Carlo Tree Search","authors":"Yuta Ishizaki, Toshinori Takayama, N. Togawa","doi":"10.1109/ICCE-Berlin47944.2019.8966146","DOIUrl":"https://doi.org/10.1109/ICCE-Berlin47944.2019.8966146","url":null,"abstract":"In this paper, we propose a route recommendation method, called P-UCT method, considering individual user’s preferences utilizing Monte-Carlo tree search. In the proposed method, we firstly extract route features based on the route recommendation history of every user and construct a route evaluator based on Support Vector Machine (SVM). After that, the method generates a random route from a start point to an end point by Monte-Carlo tree search. The route evaluator determines how well every generated route matches the user’s preferences. By repeating the evaluation, the method obtains the route, which must be closest to the user’s preferences. Experimental results demonstrate that the proposed method outperforms the existing method from the viewpoint of the average evaluation scores.","PeriodicalId":290753,"journal":{"name":"2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117160567","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}
引用次数: 1
首页 上一页 下一页 尾页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信