检索结果-内蒙古大学图书馆

Action understanding and active inference

BIOLOGICAL CYBERNETICS 2011年第1-2期104卷 137-160页

作者： Friston, Karl J. Mattout, Jeremie Kilner, James UCL Inst Neurol Wellcome Trust Ctr Neuroimaging London WC1N 3BG England INSERM U821 F-69008 Lyon France

We have suggested that the mirror-neuron system might be usefully understood as implementing Bayes-optimal perception of actions emitted by oneself or others. To substantiate this claim, we present neuronal simulations that show the same representations can prescribe motor behavior and encode motor intentions during action-observation. These simulations are based on the free-energy formulation of active inference, which is formally related to predictive coding. In this scheme, (generalised) states of the world are represented as trajectories. When these states include motor trajectories they implicitly entail intentions (future motor states). Optimizing the representation of these intentions enables predictive coding in a prospective sense. Crucially, the same generative models used to make predictions can be deployed to predict the actions of self or others by simply changing the bias or precision (i.e. attention) afforded to proprioceptive signals. We illustrate these points using simulations of handwriting to illustrate neuronally plausible generation and recognition of itinerant (wandering) motor trajectories. We then use the same simulations to produce synthetic electrophysiological responses to violations of intentional expectations. Our results affirm that a Bayes-optimal approach provides a principled framework, which accommodates current thinking about the mirror-neuron system. Furthermore, it endorses the general formulation of action as active inference.

关键词： Action-observation Mirror-neuron system Inference Precision Free-energy Perception Generative models predictive coding

来源：评论

学校读者我要写书评

暂无评论

Insights From Generative Modeling for Neural Video Compression

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2023年第8期45卷 9908-9921页

作者： Yang, Ruihan Yang, Yibo Marino, Joseph Mandt, Stephan Univ Calif Irvine Dept Comp Sci Irvine CA 92697 USA CALTECH DeepMind Pasadena CA 91125 USA

While recent machine learning research has revealed connections between deep generative models such as VAEs and rate-distortion losses used in learned compression, most of this work has focused on images. In a similar spirit, we view recently proposed neural video coding algorithms through the lens of deep autoregressive and latent variable modeling. We present these codecs as instances of a generalized stochastic temporal autoregressive transform, and propose new avenues for further improvements inspired by normalizing flows and structured priors. We propose several architectures that yield state-of-the-art video compression performance on high-resolution video and discuss their tradeoffs and ablations. In particular, we propose (i) improved temporal autoregressive transforms, (ii) improved entropy models with structured and temporal dependencies, and (iii) variable bitrate versions of our algorithms. Since our improvements are compatible with a large class of existing models, we provide further evidence that the generative modeling viewpoint can advance the neural video coding field.

关键词： Transforms Video compression Data models Image coding predictive coding Streaming media Rate-distortion Autoregressive models generative models normalizing flow variational inference video compression

来源：评论

学校读者我要写书评

暂无评论

Proposal for Infrared Atmospheric Sounding Interferometer on-board data compression

引用

JOURNAL OF APPLIED REMOTE SENSING 2015年第1期9卷

作者： Garcia-Sobrino, Joaquin Blanes, Ian Albinet, Mathieu Camarero, Roberto Serra-Sagrista, Joan Univ Autonoma Barcelona Dept Informat & Commun Engn Cerdanyola Del Valles 08193 Barcelona Spain CNES F-31401 Toulouse 9 France

The Infrared Atmospheric Sounding Interferometer (IASI) system provides infrared soundings of moisture and temperature profiles, as well as soundings of chemical components. These measurements play a key role in atmospheric chemistry, global change, and climate monitoring. The instrument, developed by a cooperating agreement between European Organisation for the Exploitation of Meteorological Satellites and Centre National d'Etudes Spatiales, is implemented on the Metop satellite series. The instrument data production rate is 45 Mb/s while the transmission rate allocated to IASI measurements is 1.5 Mb/s. It is thus necessary to implement a significant part of the IASI data processing on-board the instrument. We investigate the information statistics of IASI L0 data once the on-board processing chain is finished. We analyze order-0 entropy, and order-1, order-2 and order-3 conditional entropies, where conditional entropies assess both the spectral and the spatial joint information. According to the simple order-0 entropy, at least one bit per sample could be spared if a variable-length code was employed. We also investigate the actual performance of different lossless compression techniques on IASI L0 data. The CCSDS-123, JPEG-LS, and JPEG2000 standards, as well as M-CALIC coding technique are evaluated. Experimental results reveal that IASI Level 0 data can be coded by a compression ratio above 2.6:1. (C) 2015 Society of Photo-Optical Instrumentation Engineers (SPIE)

关键词： remote sensing compression lossless data coding predictive coding entropy analysis multi- and hyperspectral imagery CCSDS 123

来源：评论

学校读者我要写书评

暂无评论

Reversible data hiding based on histogram modification of prediction-error

引用

IMAGING SCIENCE JOURNAL 2011年第5期59卷 278-292页

作者： Lee, C-F Chen, H-L Chaoyang Univ Technol Dept Informat Management Taichung 41349 Taiwan Chaoyang Univ Technol Doctoral Program Grad Inst Informat Taichung 41349 Taiwan

This work presents a novel histogram-based reversible data hiding scheme. Although common histogram-based reversible data hiding schemes can achieve high image quality, embedding capacity is restricted because general images usually do not contain a great number of pixels with the same pixel values. To improve embedding capacity and retain low distortion, the proposed scheme uses prediction-error values, which are derived from the difference between an original image and a predictive image, instead of using the original pixels to convey a secret message. In the proposed scheme, a predictive image is generated using the mean interpolation prediction method. Since the obtained predictive image is very similar to the original image, prediction-error values are to be tended to zero. That is, a great quantity of peak points gathers around zero. The proposed scheme takes full advantage of this property to increase embedding capacity and retain slight distortion. Moreover, a threshold is used to balance the tradeoff between embedding capacity and image quality, i.e. embedding capacity in the proposed scheme is scalable. Furthermore, only a threshold is needed to record, not a large amount of information of peak and zero points, when high embedding capacity is required. Additionally, a multilevel mechanism is employed to further increase embedding capacity. Experimental results indicate that the proposed scheme is superior to other reversible schemes in terms of both image quality and embedding capacity.

关键词： reversible data hiding lossless data hiding predictive coding histogram modification

来源：评论

学校读者我要写书评

暂无评论

L_∞ constrained high-fidelity image compression via adaptive context modeling

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2000年第4期9卷 536-542页

作者： Wu, XL Bao, P Univ Western Ontario Dept Comp Sci London ON N6A 5B7 Canada Hong Kong Polytech Univ Dept Comp Hong Kong Hong Kong Peoples R China

In this paper, vee study high-fidelity image compression with a given tight L-infinity bound. We propose some practical adaptive context modeling techniques to correct prediction biases caused by quantizing prediction residues, a problem common to the existing DPCM-type predictive near-lossless image coders. By incorporating the proposed techniques into the near-lossless version of CALIC that is considered by many as state-of-the-art algorithm, me mere able to increase its PSNR by 1 dB or more and/or reduce its bit rate by 10% or more. More encouragingly, at bit rates around 1.25 bpp or higher, our method obtained competitive PSNR results against the best L-2-based wavelet coders, while obtaining much smaller L-infinity bound.

关键词： context modeling of images entropy coding near-lossless image compression predictive coding trellis quantization

来源：评论

学校读者我要写书评

暂无评论

CHANNEL ERROR PROPAGATION IN ADAPTIVE TV CODERS

引用

SIGNAL PROCESSING 1986年第2期10卷 171-183页

作者： HWANG, CJ RAO, KR Department of Electrical Engineering The University of Texas at Arlington Arlington TX 76019 U.S.A.

DPCM coders with adaptive predictors are used and compared with nonadaptive DPCM coders for processing composite color television signals. The transmission error propagation for different predictors both adaptive and fixed is investigated. Propagation of transmission noise is dependent on the type of prediction and on the location of noise, i.e., whether in a uniform region or in an active region. By introducing leak in predictor output and/or predictor function, it is shown that error propagation can be significantly reduced. The leaky predictors not only attenuate and/or terminate the channel error propagation but also improve the predictor performance based on quantitative evaluation such as peak value and mean-square-error. In order to protect the high picture quality from error propagation, an error-correcting code is needed. Der Einsatz von DPCM-Codierverfahren mit adaptivem Prädiktor bei der Verarbeitung zusammengesetzter Videosignale wird untersucht; diese Verfahren werden mit adaptiven DPCM-Codierverfahren verglichen. Insbesondere wird für verschiedene—adaptive wie nichtadaptive—Prädiktoren die Fehlerfortpflanzung bei Übertragungsstörungen untersucht. Die Fortpflanzung rauschartiger Übertragungsstörungen hängt von der Art des Prädiktors ab; auβerdem besteht eine Abhängigkeit von der Stelle, an der das Rauschen auftritt, d.h., davon, ob das Bild an dieser Stelle wenig oder stark strukturiert ist. Wie gezeigt wird, kann die Fehlerfortpflanzung durch die Einfügung einer künstlichen Dämpfung in das Ausgangssignal des Prädiktors und/oder die zur Übertragung anstehende Prädiktionsfunktion erheblich reduziert werden. Prädiktoren dieser Art dämpfen bzw. beenden nicht nur die Fortpflanzung von Übertragungsfehlern, sie verbessern darüber hinaus auch das Verhalten des Gesamtsystems, wie sich durch quantitative Messungen mit Hilfe des Kriteriums der maximalen Abweichung oder des mittleren quadratischen Fehlers nachweisen läβt. Um die hochwertige Bildqualität vor Übertr

关键词： predictive coding TV signals channel error propagation

来源：评论

学校读者我要写书评

暂无评论

Acquisition of nonlinear forward optics in generative models: Two-stage "downside-up" learning for occluded vision

引用

NEURAL NETWORKS 2011年第2期24卷 148-158页

作者： Tajima, Satohiro Watanabe, Masataka NHK Japan Broadcasting Corp Nagano Stn Nagano 3808502 Japan Univ Tokyo Grad Sch Frontier Sci Kashiwa Chiba 2778561 Japan Univ Tokyo Fac Engn Bunkyo Ku Tokyo 1130033 Japan

We propose a two-stage learning method which implements occluded visual scene analysis into a generative model, a type of hierarchical neural network with bi-directional synaptic connections. Here, top-down connections simulate forward optics to generate predictions for sensory driven low-level representation, whereas bottom-up connections function to send the prediction error, the difference between the sensory based and the predicted low-level representation, to higher areas. The prediction error is then used to update the high-level representation to obtain better agreement with the visual scene. Although the actual forward optics is highly nonlinear and the accuracy of simulated forward optics is crucial for these types of models, the majority of previous studies have only investigated linear and simplified cases of forward optics. Here we take occluded vision as an example of nonlinear forward optics, where an object in front completely masks out the object behind. We propose a two-staged learning method inspired by the staged development of infant visual capacity. In the primary learning stage, a minimal set of object basis is acquired within a linear generative model using the conventional unsupervised learning scheme. In the secondary learning stage, an auxiliary multi-layer neural network is trained to acquire nonlinear forward optics by supervised learning. The important point is that the high-level representation of the linear generative model serves as the input and the sensory driven low-level representation provides the desired output. Numerical simulations show that occluded visual scene analysis can indeed be implemented by the proposed method. Furthermore, considering the format of input to the multi-layer network and analysis of hidden-layer units leads to the prediction that whole object representation of partially occluded objects, together with complex intermediate representation as a consequence of nonlinear transformation from non-occluded to

关键词： Vision Generative model predictive coding Neural network Occlusion Learning Developmental stage

来源：评论

学校读者我要写书评

暂无评论

Study on Self-Adaptive Message Scheduler Used for the Vehicle Ad-hoc Network

引用

JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 2017年第5期33卷 1285-1305页

作者： Liu, Mingjian Tan, Guozhen Ding, Nan Zang, Fuxin Dalian Univ Technol Comp Sci Inst Dalian 116024 Peoples R China

To satisfy the strict demands of VANET (vehicular ad-hoc network) on communication rate, response speed as well as on reliability, a communication sublayer that provides reliable, real-time and qualified communication services for upper applications is constructed in VANET. By using the queuing theory and based on researching the message queues in VANET, the current study designed a scheme for optimizing the communication sublayer, and proposed mechanisms for message compression, sending-frequency self-adaption as well as for data transmission compression, which make up an easy-to-be deployed framework that provides safe, real-time and standard self-adaptive communication services for upper applications and which solve the problem that previous VANET communication is not reliable and real-time. Through simulation and real vehicle experiments,we verify that the design can well satisfy the communication requirements of VANET in terms of performance and functionality, bandwidth occupancy decrease by 47.8 % and 11.4% relatively compared with CMS (common message set) and unoptimized MD (message dispatcher) decreased,compared with CMS and unoptimized MD, the transmission frequency can decrease by 60% and 6.1%, and data compression ratio is 12.3%, thus proving the effectiveness of our scheme.

关键词： vehicle ad-hoc network communication layer queuing theory instantaneity predictive coding data compression

来源：评论

学校读者我要写书评

暂无评论

Hybrid Neural Network predictive-Wavelet Image Compression System

引用

NEUROCOMPUTING 2015年第Part3期151卷 975-984页

作者： Hussain, Abir Jaafar Al-Jumeily, Dhiya Radi, Naeem Lisboa, Paulo Liverpool John Moores Univ Sch Comp & Math Sci Liverpool L3 3AF Merseyside England Al Khawarizmi Univ Coll Al Ain U Arab Emirates

This paper considers a novel image compression technique called hybrid predictive wavelet coding. The new proposed technique combines the properties of predictive coding and discrete wavelet coding. In contrast to JPEG2000, the image data values are pre-processed using predictive coding to remove interpixel redundancy. The error values, which are the difference between the original and the predicted values, are discrete wavelet coding transformed. In this case, a nonlinear neural network predictor is utilised in the predictive coding system. The simulation results indicated that the proposed technique can achieve good compressed images at high decomposition levels in comparison to JPEG2000. (C) 2014 Elsevier B.V. All rights reserved.

关键词： Neural network predictive coding Image compression Quantisation

来源：评论

学校读者我要写书评

暂无评论

Frequency-Domain Intra Prediction Analysis and Processing for High-Quality Video coding

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2015年第5期25卷 798-811页

作者： Blasi, Saverio G. Mrak, Marta Izquierdo, Ebroul Queen Mary Univ London Sch Elect Engn & Comp Sci London E1 4NS England British Broadcasting Corp Dept Res & Dev London W12 7SB England

Most of the advances in video coding technology focus on applications that require low bitrates, for example, for content distribution on a mass scale. For these applications, the performance of conventional coding methods is typically sufficient. Such schemes inevitably introduce large losses to the signal, which are unacceptable for numerous other professional applications such as capture, production, and archiving. To boost the performance of video codecs for high-quality content, better techniques are needed especially in the context of the prediction module. An analysis of conventional intra prediction methods used in the state-of-the-art High Efficiency Video coding (HEVC) standard is reported in this paper, in terms of the prediction performance of such methods in the frequency domain. Appropriately modified encoder and decoder schemes are presented and used for this paper. The analysis shows that conventional intra prediction methods can be improved, especially for high frequency components of the signal which are typically difficult to predict. A novel approach to improve the efficiency of high-quality video coding is also presented in this paper based on such analysis. The modified encoder scheme allows for an additional stage of processing performed on the transformed prediction to replace selected frequency components of the signal with specifically defined synthetic content. The content is introduced in the signal using feature-dependent lookup tables. The approach is shown to achieve consistent gains against conventional HEVC with up to -5.2% coding gains in terms of bitrate savings.

关键词： Frequency estimation High Efficiency Video coding (HEVC) predictive coding video compression

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：