检索结果-内蒙古大学图书馆

ELEVATING SKELETON-BASED ACTION RECOGNITION WITH EFFICIENT MULTI-MODALITY SELF-SUPERVISION

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wei, Yiping Peng, Kunyu Roitberg, Alina Zhang, Jiaming Zheng, Junwei Liu, Ruiping Chen, Yufan 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

Self-supervised representation learning for human action recognition has developed rapidly in recent years. Most of the existing works are based on skeleton data while using a multi-modality setup. These works overlooked the differences in performance among modalities, which led to the propagation of erroneous knowledge between modalities while only three fundamental modalities, i.e., joints, bones, and motions are used, hence no additional modalities are explored. In this work, we first propose an Implicit Knowledge Exchange Module (IKEM) which alleviates the propagation of erroneous knowledge between low-performance modalities. Then, we further propose three new modalities to enrich the complementary information between modalities. Finally, to maintain efficiency when introducing new modalities, we propose a novel teacher-student framework to distill the knowledge from the secondary modalities into the mandatory modalities considering the relationship constrained by anchors, positives, and negatives, named relational cross-modality knowledge distillation. The experimental results demonstrate the effectiveness of our approach, unlocking the efficient use of skeleton-based multi-modality data. Source code will be made publicly available at https://***/desehuileng0o0/IKEM. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

Localization of Autonomous Vehicles in Tunnels Based on Roadside Multi-sensor Fusion

学校读者我要写书评

暂无评论

IEEE Transactions on Intelligent Vehicles

IEEE Transactions on Intelligent Vehicles 2024年 1-13页

作者： Zhu, Kongtao Chen, Shitao Shi, Jiamin Zhu, Ziyu Zheng, Nanning 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

Tunnels present significant challenges for the navigation and localization of autonomous vehicles due to the lack of GNSS signals and the presence of uniform scene textures. Current cooperative positioning strategies, which rely on RSS and ToF, are less effective in tunnel environments due to the unique electromagnetic conditions. To address this issue, a novel Vehicle-to-Infrastructure (V2I) cooperative localization methodology is introduced. The system comprises a roadside subsystem and a vehicle-side subsystem. In the roadside subsystem, data from multiple sensors is collected to calculate the vehicleâ's position. This information is then transmitted to the vehicle-side subsystem via V2I communication, where it is fused with onboard module data. A highly effective co-location process for the vehicle-side subsystem is proposed, along with an exemplary algorithm. This process resolves issues related to data dimension inconsistency, delay, and pose fusion commonly encountered during co-location. Furthermore, a comprehensive investigation and analysis explores factors that may impact positioning performance. This meticulous examination enhances the understanding of the framework and reveals its limitations. IEEE

关键词： Location

Learning Object-Centric Representation via Reverse Hierarchy Guidance

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zou, Junhong Zhu, Xiangyu Zhang, Zhaoxiang Lei, Zhen State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences Hong Kong

Object-Centric Learning (OCL) seeks to enable Neural Networks to identify individual objects in visual scenes, which is crucial for interpretable visual comprehension and reasoning. Most existing OCL models adopt auto-encoding structures and learn to decompose visual scenes through specially designed inductive bias, which causes the model to miss small objects during reconstruction. Reverse hierarchy theory proposes that human vision corrects perception errors through a top-down visual pathway that returns to bottom-level neurons and acquires more detailed information, inspired by which we propose Reverse Hierarchy Guided Network (RHGNet) that introduces a top-down pathway that works in different ways in the training and inference processes. This pathway allows for guiding bottom-level features with top-level object representations during training, as well as encompassing information from bottom-level features into perception during inference. Our model achieves SOTA performance on several commonly used datasets including CLEVR, CLEVRTex and MOVi-C. We demonstrate with experiments that our method promotes the discovery of small objects and also generalizes well on complex real-world scenes. Code will be available at https://***/r/RHGNet-6CEF. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

Self-Centering 3-DoF Feet Controller for Hands-Free Locomotion Control in Telepresence and Virtual Reality

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Memmesheimer, Raphael Lenz, Christian Schwarz, Max Schreiber, Michael Behnke, Sven The Autonomous Intelligent Systems group Computer Science Institute VI - Intelligent Systems and Robotics Lamarr Institute for Machine Learning and Artificial Intelligence Center for Robotics University of Bonn Germany

We present a novel seated feet controller for handling 3 Degree of Freedom (DoF) aimed to control locomotion for telepresence robotics and virtual reality environments. Tilting the feet on two axes yields in forward, backward and sideways motion. In addition, a separate rotary joint allows for rotation around the vertical axis. Attached springs on all joints self-center the controller. The HTC Vive tracker is used to translate the trackers' orientation into locomotion commands. The proposed self-centering feet controller was used successfully for the ANA Avatar XPRIZE competition, where a naive operator traversed the robot through a longer distance, surpassing obstacles while solving various interaction and manipulation tasks in between. We publicly provide the models of the mostly 3D-printed feet controller for reproduction. Copyright © 2024, The Authors. All rights reserved.

关键词： Biped locomotion

Comparison of cross-subject EEG emotion recognition algorithms in the BCI Controlled Robot Contest in World Robot Contest 2021

学校读者我要写书评

暂无评论

Brain Science Advances 2022年第2期8卷 142-152页

作者： Chao Tang Yunhuan Li Badong Chen Institute of Artificial Intelligence and Robotics Xi'an Jiaotong UniversityXi'an 710049China

Electroencephalogram(EEG)data depict various emotional states and reflect brain *** has been increasing interest in EEG emotion recognition in brain-computer interface systems(BCIs).In the World Robot Contest(WRC),the BCI Controlled Robot Contest successfully staged an emotion recognition technology *** types of emotions(happy,sad,and neutral)are modeled using EEG *** this study,5 methods employed by different teams are *** results reveal that classical machine learning approaches and deep learning methods perform similarly in offline recognition,whereas deep learning methods perform better in online cross-subject decoding.

关键词： electroencephalography emotion recognition online decoding cross-subject brain-computer interface

Smart Cities Using artificial intelligence (AI) and Blockchain in the Internet of Things (IoT) Era 1st

学校读者我要写书评

暂无评论

Smart Cities Using Artificial Intelligence (AI) and Blockcha...

1st International Conference on Smart Environment and Green Technologies, ICSEGT 2024

作者： Al-Turjman, Fadi Salama, Ramiz Sadikoglu, Gunay Artificial Intelligence Engineering Dept. AI and Robotics Institute Near East University Nicosia Mersin Turkey Research Center for AI and IoT Faculty of Engineering University of Kyrenia Kyrenia Mersin Turkey Department of Computer Engineering AI and Robotics Institute Research Center for AI and IoT Near East University Nicosia Mersin Turkey Near East University Nicosia Mersin Turkey

ISBN: (纸本)9783031815638

Around the world, modern cities are changing dramatically in order to install smart infrastructure, provide smart services to residents, improve accessibility for vulnerable populations, and provide a clean, sustainable, and secure environment. Numerous initiatives that address significant issues with urban transportation, health, water management, energy conservation, housing, safety, and accessibility are centered on the adoption of new technologies. To overcome these obstacles and enhance the resourcefulness, sustainability, and security of cities, it is imperative to embrace new and creative technologies, especially in light of the advancements in communication and sensor technology over the past few decades. This article gives a summary of the metropolitan issues that smart urban communities are now experiencing. It investigates how edge AI and blockchain, two recent technological innovations, might revolutionize urban intelligence. It also examines significant research initiatives in these two crucial areas of today’s smart cities, along with a variety of applications for edge AI and blockchain in smart energy and transportation. It discusses numerous uses of blockchain in the context of secure vehicle-to-vehicle trade and communication, along with a range of methods for processing speed detection, counting, and identification to lessen traffic congestion. The purpose of this review article is to set the stage for next research on applying blockchain technology and edge artificial intelligence to smarter city applications. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Sustainable city

Pedestrians and Robots: A Novel Dataset for Learning Distinct Social Navigation Forces

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Agrawal, Subham Ostermann-Myrau, Nico Dengler, Nils Bennewitz, Maren Humanoid Robots Lab University of Bonn Germany Lamarr Institute for Machine Learning and Artificial Intelligence The Center for Robotics Bonn Germany

The increasing use of robots in human-centric public spaces such as shopping malls, sidewalks, and hospitals, requires understanding of how pedestrians respond to their presence. However, existing research lacks comprehensive datasets that capture the full range of pedestrian behaviors, e.g., including avoidance, neutrality, and attraction in the presence of robots. Such datasets can be used to effectively learn models capable of accurately predicting diverse responses of pedestrians to robot presence, which are crucial for advancing robot navigation strategies and optimizing pedestrian-aware motion planning. In this paper, we address these challenges by collecting a novel dataset of pedestrian motion in two outdoor locations under three distinct conditions, i.e., no robot presence, a stationary robot, and a moving robot. Thus, unlike existing datasets, ours explicitly encapsulates variations in pedestrian behavior across the different robot conditions. Using our dataset, we propose a novel Neural Social Robot Force Model (NSRFM), an extension of the traditional Social Force Model that integrates neural networks and robot-induced forces to better predict pedestrian behavior in the presence of robots. We validate the NSRFM by comparing its generated trajectories on different real-world datasets. Furthermore, we implemented it in simulation to enable the learning and benchmarking of robot navigation strategies based on their impact on pedestrian movement. Our results demonstrate the model’s effectiveness in replicating real-world pedestrian reactions and its its utility in developing, evaluating, and benchmarking social robot navigation algorithms. Copyright © 2025, The Authors. All rights reserved.

关键词： Social robots

MOTPose: Multi-object 6D Pose Estimation for Dynamic Video Sequences using Attention-based Temporal Fusion

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Periyasamy, Arul Selvam Behnke, Sven the Autonomous Intelligent Systems group Computer Science Institute VI – Intelligent Systems and Robotics the Center for Robotics the Lamarr Institute for Machine Learning and Artificial Intelligence University of Bonn Germany

Cluttered bin-picking environments are challenging for pose estimation models. Despite the impressive progress enabled by deep learning, single-view RGB pose estimation models perform poorly in cluttered dynamic environments. Imbuing the rich temporal information contained in the video of scenes has the potential to enhance models’ ability to deal with the adverse effects of occlusion and the dynamic nature of the environments. Moreover, joint object detection and pose estimation models are better suited to leverage the co-dependent nature of the tasks for improving the accuracy of both tasks. To this end, we propose attention-based temporal fusion for multi-object 6D pose estimation that accumulates information across multiple frames of a video sequence. Our MOTPose method takes a sequence of images as input and performs joint object detection and pose estimation for all objects in one forward pass. It learns to aggregate both object embeddings and object parameters over multiple time steps using cross-attention-based fusion modules. We evaluate our method on the physically-realistic cluttered bin-picking dataset SynPick and the YCB-Video dataset and demonstrate improved pose estimation accuracy as well as better object detection accuracy. © 2024, CC BY.

关键词： Object detection

Safety-Critical Scenario Generation by Causal Influence Detection 27

学校读者我要写书评

暂无评论

Safety-Critical Scenario Generation by Causal Influence Dete...

27th IEEE International Conference on Intelligent Transportation Systems, ITSC 2024

作者： Yang, Yibing Zhang, Chi Xu, Linhai Ma, Shuangxun Li, Li Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Shaanxi Xi'an China School of Future Transportation Chang'an University Xi'an710064 China Tsinghua University BNRist Department of Automation Beijing100084 China

ISBN: (纸本)9798331505929

Appropriate behavior intervention for a specific background vehicle at the right moment is a crucial step in identifying safety-critical scenarios. Existing methods generate scenarios by estimating the data distribution, but when noise interventions are present, the emergence of safety-critical scenarios becomes colinearity with the noise interventions, potentially resulting in unnecessary interventions and unrealistic vehicle behaviors. To address this, we propose to integrate the causal relationship between the background vehicle behavior and the safety status of ego-vehicle as a prior into scenario generation. In this notion, we present a reinforcement learning framework with a causal influence detection module (CausalID). Specifically, we employ conditional mutual information to quantify the causal influence between vehicle sequential behaviors, and intervene on the background vehicle's behavior when the causal influence exceeds a certain threshold. Additionally, to prevent unrealistic collision-oriented driving, we sample the final intervention action from a set of candidate actions according to their probabilities of occurring in natural driving scenarios. Our experiments in 3-lane highway scenarios with multiple vehicles validate the effectiveness of the proposed framework. © 2024 IEEE.

关键词： Vehicle safety