检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Chen, Long Song, Xiaotian Song, Andy Chen, BaDong Lv, Jiancheng Sun, Yanan School of Computer Science Sichuan University China School of Computing Technologies Royal Melbourne Institute Technology University Australia Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China

Spiking Large Language Models have been shown as a good alternative to LLMs in various scenarios. Existing methods for creating Spiking LLMs, i.e., direct training and ANN-SNN conversion, often suffer from performance degradation and relatively high computational costs. To address these issues, we propose a novel Fast ANN-SNN conversion strategy (FAS) that transforms LLMs into spiking LLMs in two stages. The first stage employs a full-parameter fine-tuning of pre-trained models, so it does not need any direct training from scratch. The second stage introduces a coarse-to-fine calibration method to reduce conversion errors and improve accuracy. Our experiments on both language and vision-language tasks across four different scales of LLMs demonstrate that FAS can achieve state-of-the-art performance yet with significantly reduced inference latency and computational costs. For example, FAS only takes 8 timesteps to achieve an accuracy of 3% higher than that of the OPT-7B model, while reducing energy consumption by 96.63%. Copyright © 2025, The Authors. All rights reserved.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

NUSense: Robust Soft Optical Tactile Sensor

arXiv

引用

arXiv 2024年

作者： Yergibay, Madina Mussin, Tleukhan Seitzhan, Saltanat Kenzhebek, Daryn Kappassov, Zhanat Soh, Harold Taunyazov, Tasbolat Robotics Department Institute of Smart Systems and Artificial Intelligence Nazarbayev University Astana Kazakhstan Dept. of Computer Science National University of Singapore Singapore

While most tactile sensors rely on measuring pressure, insights from continuum mechanics suggest that measuring shear strain provides critical information for tactile sensing. In this work, we introduce an optical tactile sensing principle based on shear strain detection. A silicone rubber layer, dyed with color inks, is used to quantify the shear magnitude of the sensing layer. This principle was validated using the NUSense camera-based tactile sensor. The wide-angle camera captures the elongation of the soft pad under mechanical load, a phenomenon attributed to the Poisson effect. The physical and optical properties of the inked pad are essential and should ideally remain stable over time. We tested the robustness of the sensor by subjecting the outermost layer to multiple load cycles using a robot arm. Additionally, we discussed potential applications of this sensor in force sensing and contact localization. Copyright © 2024, The Authors. All rights reserved.

关键词： Silicones

来源：评论

学校读者我要写书评

暂无评论

Function-oriented optimization design method for underactuated tendon-driven humanoid prosthetic hand

引用

Frontiers of Mechanical Engineering 2022年第3期17卷 75-89页

作者： Yue ZHENG Xiangxin LI Lan TIAN Guanglin LI CAS Key Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institutes of Advanced Technology(SIAT)Chinese Academy of Sciences(CAS)Shenzhen518055China Shenzhen College of Advanced Technology University of Chinese Academy of SciencesShenzhen518055China SIAT Branch Shenzhen Institute of Artificial Intelligence and Robotics for SocietyShenzhen518055China Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen518055China

The loss of hand functions in upper limb amputees severely restricts their mobility in daily *** a humanoid prosthetic hand would be an effective way of restoring lost hand *** a prosthetic hand design,replicating the natural and dexterous grasping functions with a few actuators remains a big *** this study,a function-oriented optimization design(FOD)method is proposed for the design of a tendon-driven humanoid prosthetic *** optimization function of different functional conditions of full-phalanx contact,total contact force,and force isotropy was constructed based on the kinetostatic model of a prosthetic finger for the evaluation of grasping *** a genetic algorithm,the optimal geometric parameters of the prosthetic finger could be determined for specific functional *** results reveal that the structure of the prosthetic finger is significantly different when designed for different functional requirements and grasping target sizes.A prosthetic finger was fabricated and tested with grasping *** mean absolute percentage error between the theoretical value and the experimental result is less than 10%,demonstrating that the kinetostatic model of the prosthetic finger is effective and makes the FOD method *** study suggests that the FOD method enables the systematic evaluation of grasping performance for prosthetic hands in the design stage,which could improve the design efficiency and help prosthetic hands meet the design requirements.

关键词： function-oriented tendon driven prosthetic hand optimization humanoid underactuated

来源：评论

学校读者我要写书评

暂无评论

CPL-CLIP: Compound Prompt Learning for Flexible-Modal Face Anti-Spoofing

CPL-CLIP: Compound Prompt Learning for Flexible-Modal Face A...

引用

IEEE International Joint Conference on Biometrics (IJCB)

作者： Ying Zhang Xiangyu Zhu Ajian Liu Xun Lin Jun Wan Jing Zhang Zhen Lei State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences School of Artificial Intelligence University of Chinese Academy of Sciences School of Computer Science and Engineering Beihang University School of Computer Science and Engineering Macau University of Science and Technology Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences

ISBN: (数字)9798350364132

ISBN: (纸本)9798350364149

Face anti-spoofing (FAS) is pivotal in safeguarding the integrity of face recognition systems. Flexible-modal FAS utilizes multi-modal data and trains a unified model adaptable to any single-modal testing scenario. This innovation addresses the shortcomings of conventional multi-modal FAS approaches, which typically demand separate model training and deployment for each modality. However, existing flexible-modal FAS approaches activate specific network branches based on the modality of the tested sample. This not only increases the model’s parameters but also necessitates the provision of the image’s modality for testing, thereby constraining deployment flexibility. To address the issue, we present Compound Prompt Learning CLIP (CPL-CLIP), a novel method for flexible-modal FAS. This approach capitalizes on a learned textual prompt that is nearly independent of modality, thus bolstering class-based classification across arbitrary modalities. Specifically, our CPL-CLIP introduces a Dual-Branch Prompt (DBP), consisting of class and modal prompts that describe and guide classification, where each prompt is composed of learnable vectors and fixed templates. To further render the class prompt as modality-agnostic as possible, a Cosine Similarity Loss (CSL) is proposed to facilitate the maximal separation of the class prompt from the modality prompt. With only the class prompt utilized during testing, CPL-CLIP enables deployment in diverse modal testing scenarios without the necessity of the test image’s modality to be known. Extensive experiments demonstrate CPL-CLIP’s superiority over existing methods on several flexible-modal FAS benchmarks.

关键词： Training Adaptation models Technological innovation Accuracy Face recognition Benchmark testing Vectors Robustness Data models Compounds

来源：评论

学校读者我要写书评

暂无评论

Diffusion model with cross attention as an inductive bias for disentanglement 24

Diffusion model with cross attention as an inductive bias fo...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Tao Yang Cuiling Lan Yan Lu Nanning Zheng National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University China Microsoft Research Asia

ISBN: (纸本)9798331314385

Disentangled representation learning strives to extract the intrinsic factors within the observed data. Factoring these representations in an unsupervised manner is notably challenging and usually requires tailored loss functions or specific structural designs. In this paper, we introduce a new perspective and framework, demonstrating that diffusion models with cross-attention itself can serve as a powerful inductive bias to facilitate the learning of disentangled representations. We propose to encode an image into a set of concept tokens and treat them as the condition of the latent diffusion model for image reconstruction, where cross attention over the concept tokens is used to bridge the encoder and the U-Net of the diffusion model. We analyze that the diffusion process inherently possesses the time-varying information bottlenecks. Such information bottlenecks and cross attention act as strong inductive biases for promoting disentanglement. Without any regularization term in the loss function, this framework achieves superior disentanglement performance on the benchmark datasets, surpassing all previous methods with intricate designs. We have conducted comprehensive ablation studies and visualization analyses, shedding a light on the functioning of this model. We anticipate that our findings will inspire more investigation on exploring diffusion model for disentangled representation learning towards more sophisticated data analysis and understanding.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Vehicle Trajectory Prediction with Soft Behavior Constraints

Vehicle Trajectory Prediction with Soft Behavior Constraints

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Ke Ye Sanping Zhou Miao Kang Jingwen Fu Nanning Zheng National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Shaanxi China

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Trajectory prediction plays a crucial role in autonomous driving, but it is challenging due to the multi-modal nature of future trajectories. Behavior information is frequently employed to capture more diverse modalities of future trajectories. Traditional behavior information is typically hard-encoded, which is often inaccurate and inadequate for reflecting future multimodality. Therefore, we introduce the concept of soft vehicle behavior, which is represented as a probability distribution over a predefined comprehensive set of behaviors. This approach allows for a more rational depiction of vehicle behavior and captures potential future driving modalities. Based on it, we propose a new soft-behavior-constrained vehicle trajectory prediction framework. The framework consists of a backbone and a lightweight and plug-and-play behavior prediction module, which is used to imbue soft behavior constraints to assist in representation learning. We integrated the behavior prediction module into five representative trajectory predictors and achieved improvements of at least 4.2% in minFDE(K=5) on the nuScenes dataset and 0.5% in minFDE(K=6) on the Argoverse 1 motion forecasting dataset. These universal increments prove the effectiveness and generalizability of soft behavior constraints in vehicle trajectory prediction.

关键词： Representation learning Space vehicles Philosophical considerations Pedestrians Buildings Predictive models Probability distribution Trajectory Forecasting Intelligent robots

来源：评论

学校读者我要写书评

暂无评论

Dynamic Obstacle Avoidance of An Omnidirectional USV

Dynamic Obstacle Avoidance of An Omnidirectional USV

引用

IEEE International Conference on Real-time Computing and robotics (RCAR)

作者： Ziyou Wang Lianxin Zhang Yang Lin Huihuan Qian School of Science and Engineering The Chinese University of Hong Kong Shenzhen Shenzhen Institute of Artificial Intelligence and Robotics for Society The Chinese University of Hong Kong Shenzhen

ISBN: (数字)9798350372601

ISBN: (纸本)9798350372618

This paper addresses the significance of dynamic collision avoidance for robots operating in environments with numerous obstacles. We propose a path planning algorithm based on RRT for omnidirectional unmanned surface vessels (USV) to navigate around static and dynamic obstacles. Our algorithm employs linear regression to predict future positions of dynamic obstacles based on their current states, including positions and velocities. These predicted positions are then incorporated into the RRT * path planning process. The algorithm aims to achieve a prediction after a fixed period, ensuring real-time collision avoidance. In addition, simulations explore various map and prediction times to evaluate the algorithm’s performance, ultimately determining that a prediction time of 0.5 s is optimal. This experiment underscores the algorithm’s capability to navigate and respond to complex environments effectively.

关键词： Navigation Heuristic algorithms Predictive models Prediction algorithms Real-time systems Path planning Collision avoidance

来源：评论

学校读者我要写书评

暂无评论

Voxel Mamba: Group-Free State Space Models for Point Cloud based 3D Object Detection

arXiv

引用

arXiv 2024年

作者： Zhang, Guowen Fan, Lue He, Chenhang Lei, Zhen Zhang, Zhaoxiang Zhang, Lei The Hong Kong Polytechnic University Hong Kong Center for Artificial Intelligence and Robotics HKISI CAS China Institute of Automation Chinese Academy of Sciences China

Serialization-based methods, which serialize the 3D voxels and group them into multiple sequences before inputting to Transformers, have demonstrated their effectiveness in 3D object detection. However, serializing 3D voxels into 1D sequences will inevitably sacrifice the voxel spatial proximity. Such an issue is hard to be addressed by enlarging the group size with existing serialization-based methods due to the quadratic complexity of Transformers with feature sizes. Inspired by the recent advances of state space models (SSMs), we present a Voxel SSM, termed as Voxel Mamba, which employs a group-free strategy to serialize the whole space of voxels into a single sequence. The linear complexity of SSMs encourages our group-free design, alleviating the loss of spatial proximity of voxels. To further enhance the spatial proximity, we propose a Dual-scale SSM Block to establish a hierarchical structure, enabling a larger receptive field in the 1D serialization curve, as well as more complete local regions in 3D space. Moreover, we implicitly apply window partition under the group-free framework by positional encoding, which further enhances spatial proximity by encoding voxel positional information. Our experiments on Waymo Open Dataset and nuScenes dataset show that Voxel Mamba not only achieves higher accuracy than state-of-the-art methods, but also demonstrates significant advantages in computational efficiency. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

LPHD: Line Perspective Hexagonal Screw Detection 42

LPHD: Line Perspective Hexagonal Screw Detection

引用

42nd Chinese Control Conference, CCC 2023

作者： Ma, Jun Li, Wenjie Liu, Xiangbei He, Cheng Liu, Zhidong Guo, Xian State Grid Tianjin Electric Power Company Tianjin China Institute of Robotics and Automatic Information System Nankai University College of Artificial Intelligence Tianjin China Automobile and Rail Transportation School Tianjin Sino-German University of Applied Sciences Tianjin China

ISBN: (纸本)9789887581543

In industrial production, hexagonal screws are widely used as fasteners, and it is crucial to have fast and accurate inspection of these screws for automated screwing processes. However, this remains a challenge due to various factors such as environmental distractions, lighting, shadows, and screw positioning. To address this, the Line Perspective Hexagonal Screw Detection (LPHD) algorithm has been proposed in this paper, which introduces four innovative techniques: canny value, hexagonal base, perspective correction, and double boundary. The LPHD algorithm starts by feeding the image captured by the robot camera into the YOLO model to obtain the bounding box of the hexagonal screw. Then, the result is subjected to canny edge detection and LSD/FLD line detection. Since the head of a hexagonal screw is a regular hexagonal prism, only three edges are needed to determine a hexagonal contour, which is referred to as the hexagonal base. Perspective correction is proposed to eliminate the camera's perspective effect. If multiple contours are found, a combination of canny value and double boundary is used to identify the optimal result. Ultimately, the LPHD algorithm is applied to find the hexagonal contour in this box. This paper also compares the efficiency and accuracy of LPHD with Mask R-CNN. The results show that LPHD is nearly three times more efficient than the Mask R-CNN algorithm with approximately the same accuracy. Therefore, the LPHD algorithm is a promising solution for detecting hexagonal screws in industrial settings. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Skeleton-Based Human Action Recognition with Noisy Labels

arXiv

引用

arXiv 2024年

作者： Xu, Yi Peng, Kunyu Wen, Di Liu, Ruiping Zheng, Junwei Chen, Yufan Zhang, Jiaming Roitberg, Alina Yang, Kailun Stiefelhagen, Rainer Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology Germany Institute for Artificial Intelligence University of Stuttgart Germany School of Robotics Hunan University China National Engineering Research Center of Robot Visual Perception and Control Technology Hunan University China

Understanding human actions from body poses is critical for assistive robots sharing space with humans in order to make informed and safe decisions about the next interaction. However, precise temporal localization and annotation of activity sequences is time-consuming and the resulting labels are often noisy. If not effectively addressed, label noise negatively affects the model’s training, resulting in lower recognition quality. Despite its importance, addressing label noise for skeleton-based action recognition has been overlooked so far. In this study, we bridge this gap by implementing a framework that augments well-established skeleton-based human action recognition methods with label-denoising strategies from various research areas to serve as the initial benchmark. Observations reveal that these baselines yield only marginal performance when dealing with sparse skeleton data. Consequently, we introduce a novel methodology, NoiseEraSAR, which integrates global sample selection, co-teaching, and Cross-Modal Mixture-of-Experts (CM-MOE) strategies, aimed at mitigating the adverse impacts of label noise. Our proposed approach demonstrates better performance on the established benchmark, setting new state-of-the-art standards. The source code for this study is accessible at https://***/xuyizdby/NoiseEraSAR. Copyright © 2024, The Authors. All rights reserved.

关键词： Musculoskeletal system

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：