检索结果-内蒙古大学图书馆

Spatial-Temporal Graph U-Net for Skeleton-Based Human Motion Infilling 25

学校读者我要写书评

暂无评论

Spatial-Temporal Graph U-Net for Skeleton-Based Human Motion...

25th IEEE International Conference on Industrial Technology, ICIT 2024

作者： Xu, Leiyang Wang, Qiang Yang, Chenguang Harbin Institute Of Technology Department Of Control Science And Engineering Harbin China University Of Liverpool Department Of Computer Science LiverpoolL69 3BX United Kingdom

ISBN: (纸本)9798350340266

Motion infilling is a fundamental and challenging research field in human motion modeling and analysis, which aims to generate natural and visually coherent transitions to fill in missing motion frames based on the start and end motion sequences. However, most current methods ignore the spatial structure formed by joints, which may lose some spatial information. This work proposes a novel spatiotemporal graph U-Net that supports flexible inputs for skeleton-based motion infilling. We apply spatiotemporal graph convolutional layers, skeleton pooling layers, and skeleton unpooling layers to extract spatial and temporal features in the motion sequence. At the same time, we use the U-Net structure to integrate the information in the start and end motion sequences. In addition, the generative adversarial mechanism is introduced to ensure the generated skeleton poses are smooth and natural. We conduct experiments on two motion datasets, including one large-scale public dataset and one self-built dataset. The model inputs are joint quaternions or joint coordinates. Experimental results show that our method can improve the performance of skeleton-based motion infilling and achieve state-of-the-art results when using joint coordinates as model input. © 2024 IEEE.

关键词： Generative adversarial networks

HGO-YOLO: Advancing Anomaly Behavior Detection with Hierarchical Features and Lightweight Optimized Detection

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zheng, Qizhi Luo, Zhongze Guo, Meiyan Wang, Xinzhu Wu, Renqimuge Meng, Qiu Dong, Guanghui College of Computer and Control Engineering Northeast Forestry University Harbin150040 China

Accurate and real-time object detection is crucial for anomaly behavior detection, especially in scenarios constrained by hardware limitations, where balancing accuracy and speed is essential for enhancing detection performance. This study proposes a model called HGO-YOLO, which integrates the HGNetv2 architecture into YOLOv8. This combination expands the receptive field and captures a wider range of features while simplifying model complexity through GhostConv. We introduced a lightweight detection head, OptiConvDetect, which utilizes parameter sharing to construct the detection head effectively. Evaluation results show that the proposed algorithm achieves a mAP@0.5 of 87.4% and a recall rate of 81.1%, with a model size of only 4.6 MB and a frame rate of 56 FPS on the CPU. HGO-YOLO not only improves accuracy by 3.0% but also reduces computational load by 51.69% (from 8.9 GFLOPs to 4.3 GFLOPs), while increasing the frame rate by a factor of 1.7. Additionally, real-time tests were conducted on Raspberry Pi4 and NVIDIA platforms. These results indicate that the HGO-YOLO model demonstrates superior performance in anomaly behavior detection. © 2025, CC BY.

关键词： Object detection

Research of interleaved hysteresis control based on soft-switching power amplifier

学校读者我要写书评

暂无评论

Research of interleaved hysteresis control based on soft-swi...

China International Electrical and Energy Conference (CIEEC)

作者： Guangjuan Qiu Xinyue Zhang Jian Wei Jingxin Chen Peng E College of Computer and Control Engineering Northeast Forestry University Harbin China

ISBN: (数字)9798350359558

ISBN: (纸本)9798350359565

This paper delves into the interleaved hysteresis control method based on soft-switching power amplifiers. Firstly, through modeling and analysis, the basic topology and mode transition of the soft-switching power amplifier are clearly defined, revealing the conditions for implementing soft-switching. Secondly, a strategy for interleaved soft-switching hysteresis control is proposed, detailing its working principle, and the frequency range of the interleaved hysteresis loop is obtained through analysis. In order to optimize the design of the filter, this paper also deeply analyzes the design methods and parameter selection of the filter inductance, providing a theoretical foundation for the implementation of the power amplifier. The research results show that the proposed interleaved soft-switching hysteresis control method can reduce switch losses and electromagnetic interference, thereby improving the efficiency and reliability of the power amplifier. This study provides beneficial theoretical support and practical guidance for the application of high-performance motor drive systems.

关键词： Motor drives Circuits Power amplifiers Switches Reliability theory Topology Magnetic losses

EU Battery Regulation and the Role of Blockchain in Creating Digital Battery Passports

学校读者我要写书评

暂无评论

EU Battery Regulation and the Role of Blockchain in Creating...

Blockchain, Robotics and AI for Networking Security Conference (BRAINS)

作者： Alberto Butera Valentina Gatteschi Department of Computer and Control Engineering Politecnico di Torino Torino Italy

ISBN: (数字)9798350367843

ISBN: (纸本)9798350367850

In response to concerns about climate change and environmental sustainability, there is a global emphasis on innovative solutions to reduce emissions and optimize resource use. A key initiative is the introduction of digital passports for batteries, with the primary goal of improving the circular economy. The European Commission has taken a pioneering role by publishing regulations, laying the groundwork for digital passports and standardization. This article has a twofold aim: first, with the aim of supporting the adoption of batteries passports, it provides an overview of current initiatives and regulations in the context of electric vehicle batteries; secondly, it examines the potential of blockchain technology in implementing digital passports, in order to demonstrate its compatibility with recent regulations, such as the one proposed by the European Commission.

关键词： Protocols Publishing Green products Europe Prototypes Regulation Batteries Blockchains Proposals Standards

A Review of Findings from Neuroscience and Cognitive Psychology as Possible Inspiration for the Path to Artificial General Intelligence

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Leon, Florin Department of Computer Science and Engineering Faculty of Automatic Control and Computer Engineering "Gheorghe Asachi" Technical University of Iași Romania

This review aims to contribute to the quest for artificial general intelligence by examining neuroscience and cognitive psychology methods for potential inspiration. Despite the impressive advancements achieved by deep learning models in various domains, they still have shortcomings in abstract reasoning and causal understanding. Such capabilities should be ultimately integrated into artificial intelligence systems in order to surpass data-driven limitations and support decision making in a way more similar to human intelligence. This work is a vertical review that attempts a wide-ranging exploration of brain function, spanning from lower-level biological neurons, spiking neural networks, and neuronal ensembles to higher-level concepts such as brain anatomy, vector symbolic architectures, cognitive and categorization models, and cognitive architectures. The hope is that these concepts may offer insights for solutions in artificial general intelligence. © 2024, CC BY-NC-SA.

关键词： Decision making

Toward a Realistic Benchmark for Out-of-Distribution Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Recalcati, Pietro Garcea, Fabio Piano, Luca Lamberti, Fabrizio Morra, Lia Department of Control and Computer Engineering Politecnico di Torino Torino Italy

Deep neural networks are increasingly used in a wide range of technologies and services, but remain highly susceptible to out-of-distribution (OOD) samples, that is, drawn from a different distribution than the original training set. A common approach to address this issue is to endow deep neural networks with the ability to detect OOD samples. Several benchmarks have been proposed to design and validate OOD detection techniques. However, many of them are based on far-OOD samples drawn from very different distributions, and thus lack the complexity needed to capture the nuances of real-world scenarios. In this work, we introduce a comprehensive benchmark for OOD detection, based on ImageNet and Places365, that assigns individual classes as in-distribution or out-of-distribution depending on the semantic similarity with the training set. Several techniques can be used to determine which classes should be considered in-distribution, yielding benchmarks with varying properties. Experimental results on different OOD detection techniques show how their measured efficacy depends on the selected benchmark and how confidence-based techniques may outperform classifier-based ones on near-OOD samples. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep neural networks

A Hybrid Deep Learning Framework for Accurate Prediction of Significant Wave Height in the North Pacific Ocean

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Wang, Jianhui Cui, Zhendong School of Computer and Control Engineering Yantai University Yantai264000 China

Significant wave height (WVHT) is one of the important parameters applied in the field of ocean engineering. Accurate predictions of WVHT can help improve wave energy conversion efficiency, coastal facility management, and disaster prevention and mitigation. However, extracting features from complex ocean parameters to accurately predict wave heights remains a challenge. A hybrid deep learning framework is proposed for precise prediction of WVHT, which combining the complementary strengths of deep learning architectures and gradient boosting methods. Incorporating an attention mechanism (AM), a convolutional neural network (CNN), and a bidirectional long and short-term memory network (BiLSTM), the framework is designed to deeply learn the information contained in historical data. Furthermore, the extreme gradient boosting (XGBoost) algorithm is integrated using a multi-model fusion strategy based on weighted ensemble techniques. First, input features are selected using Pearson's correlation coefficient and XGBoost feature importance score. The feature set is then fed into AM-based CNN-BiLSTM (CBLA) and XGBoost for training to predict WVHT. Finally, the predictions of the above models are fused to produce the final predictions. We predict the WVHT at three stations in the North Pacific at 1, 2, 4, and 6 hours in advance. The results show that the prediction performance and robustness of hybrid framework outperforms that of CBLA and XGBoost, and significantly outperforms that of other benchmark models. © 2024, The Authors. All rights reserved.

关键词： Forecasting

BONES: a benchmark for neural estimation of shapley values

学校读者我要写书评

暂无评论

BONES: a benchmark for neural estimation of shapley values

International Conference on Application of Information and Communication Technologies (AICT)

作者： Davide Napolitano Luca Cagliero Department of Control and Computer Engineering Politecnico di Torino Torino Italy

ISBN: (数字)9798350387537

ISBN: (纸本)9798350387544

Shapley Values are concepts established for eXplainable AI. They are used to explain black-box predictive models by quantifying the features’ contributions to the model’s outcomes. Since computing the exact Shapley Values is known to be computationally intractable on real-world datasets, neural estimators have emerged as alternative, more scalable approaches to get approximated Shapley Values estimates. However, experiments with neural estimators are currently hard to replicate as algorithm implementations, explainer evaluators and results visualizations are neither standardized nor promptly usable. To bridge this gap, we present BONES, a new benchmark focused on neural estimation of Shapley Value. It provides researchers with a suite of state-of-the-art neural and traditional estimators, a set of commonly used benchmark datasets, ad hoc modules for training black-box models, as well as specific functions to easily compute the most popular evaluation metrics and visualize results. The purpose is to simplify XAI model usage, evaluation, and comparison. In this paper, we showcase BONES results and visualizations for XAI model benchmarking on both tabular and image data. The open-source library is available at the following link: https://***/DavideNapolitano/BONES.

关键词： Training Measurement Computational modeling Estimation Data visualization Closed box Benchmark testing Predictive models Bones Libraries

BONES: a Benchmark fOr Neural Estimation of Shapley values

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Napolitano, Davide Cagliero, Luca Department of Control and Computer Engineering Politecnico di Torino Torino Italy

Shapley Values are concepts established for eXplainable AI. They are used to explain black-box predictive models by quantifying the features’ contributions to the model’s outcomes. Since computing the exact Shapley Values is known to be computationally intractable on real-world datasets, neural estimators have emerged as alternative, more scalable approaches to get approximated Shapley Values estimates. However, experiments with neural estimators are currently hard to replicate as algorithm implementations, explainer evaluators and results visualizations are neither standardized nor promptly usable. To bridge this gap, we present BONES, a new benchmark focused on neural estimation of Shapley Value. It provides researchers with a suite of state-of-the-art neural and traditional estimators, a set of commonly used benchmark datasets, ad hoc modules for training black-box models, as well as specific functions to easily compute the most popular evaluation metrics and visualize results. The purpose is to simplify XAI model usage, evaluation, and comparison. In this paper, we showcase BONES results and visualizations for XAI model benchmarking on both tabular and image data. The opensource library is available at the following link: https://***/DavideNapolitano/BONES. © 2024, CC BY.

关键词： Benchmarking