检索结果-内蒙古大学图书馆

6th ACM International Conference on Multimedia in Asia, MMAsia 2024

作者： Wang, Zhiyuan Yang, Cong Zhang, Yulu Boukhers, Zeyd Sui, Wei Ji, Yi Liu, Chunping School of Computer Science and Techonology Soochow University Suzhou China School of Future Science and Engineering Soochow University Suzhou China Chery Hangzhou China Fraunhofer Institute for Applied Information Technology FIT Sankt Augustin Germany Horizon Robotics Beijing China

ISBN: (纸本)9798400712739

Recent advancements in skeleton extraction have significantly improved the process by simplifying the skeleton regression task into graph component detection. Despite the advancements in skeleton topology, accuracy in detailing skeletal parts remains challenging, with specific issues such as jagged edges in high-resolution images. This paper identifies the limitations of current detection models that can adapt during the decomposition and reconstruction phases, which impacts the overall precision of the extraction. In response, we propose an approach that revises the primary focus of the detection tasks. Inspired by the success of pixel-wise binary classification methods, we propose a gradual transition in focus from a coordinate localization regression task to a classification task of predicting points during the training process. This transition can be achieved by adjusting the number of object queries in the Transformer model. Theoretical and experimental evaluations validate the effectiveness of our approach. Our method yields significant improvements in performance over the baseline across various shape and image datasets (e.g., 0.836 vs. 0.826 for BlumNet on the SK1491 dataset). © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Musculoskeletal system

来源：评论

学校读者我要写书评

暂无评论

Exploiting Cross-Modal Cost Volume for Multi-sensor Depth Estimation 17th

Exploiting Cross-Modal Cost Volume for Multi-sensor Depth E...

引用

17th Asian Conference on computer Vision, ACCV 2024

作者： Kim, Janghyun Shin, Ukcheol Heo, Seokyong Park, Jinsun Department of Information Convergence Engineering Pusan National University Busan Korea Republic of Robotics Institute Carnegie Mellon University Pittsburgh United States School of Computer Science and Engineering Pusan National University Busan Korea Republic of

ISBN: (纸本)9789819609680

Single-modal depth estimation has shown steady improvement over the years. However, relying solely on a single imaging sensor such as RGB and near-infrared (NIR) cameras can result in unreliable and erroneous depth estimation, particularly in challenging lighting conditions such as low-light or sudden lighting change scenarios. Thereby, several approaches have leveraged multiple sensors for robust depth estimation. However, the effective fusion method that maximally utilizes multi-modal sensor information still requires further investigation. With this in mind, we propose a multi-modal cost volume fusion strategy with cross-modal attention, incorporating information from both cross-spectral and single-modality pairs. Our method initially constructs low-level cost volumes that consist of modality-specific (ı.e., single modality) and modality-invariant (ı.e., cross-spectral) volumes from multi-modal sensors. These cost volumes are then gradually fused using bidirectional cross-modal fusion and unidirectional LiDAR fusion to generate a multi-sensory cost volume. Furthermore, we introduce a straightforward domain gap reduction approach to learn modality-invariant features and depth refinement techniques through cost volume-guided propagation. Experimental results demonstrate that our method achieves SOTA (State-of-the-Art) performance under diverse environmental changes. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Depth perception

来源：评论

学校读者我要写书评

暂无评论

Rapid Assessment of MANET Reliability Trends

Rapid Assessment of MANET Reliability Trends

引用

International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM)

作者： Vladimir Gvozdev Aliya Davlieva Robert Galimov Department of Computer Science and Robotics Ufa University of Science and Technology Ufa Russia

The topological features of distributed infocommunication networks are among the key factors affecting such information safety characteristics as stability; reliability and resistance to failures and attacks. In the framework of this work, the postulate is postulated that the consequences of the security threats implementation are disconnections between the Wireless Mobile ad hoc network nodes. The metric characteristic of the consequences is a change in the system reliability indicators. The paper substantiates the possibility of using well-known methods for identifying trends to build estimates of the integral security characteristics of the Wireless Mobile ad hoc network; apparatus of order statistics; statistic indexes; linguistic variables.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DeTrack: In-model Latent Denoising Learning for Visual Object Tracking 38

DeTrack: In-model Latent Denoising Learning for Visual Objec...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Zhou, Xinyu Li, Jinglun Hong, Lingyi Jiang, Kaixun Guo, Pinxue Ge, Weifeng Zhang, Wenqiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China Shanghai Engineering Research Center of AI & Robotics Academy for Engineering and Technology Fudan University Shanghai China

Previous visual object tracking methods employ image-feature regression models or coordinate autoregression models for bounding box prediction. Image-feature regression methods heavily depend on matching results and do not utilize positional prior, while the autoregressive approach can only be trained using bounding boxes available in the training set, potentially resulting in suboptimal performance during testing with unseen data. Inspired by the diffusion model, denoising learning enhances the model's robustness to unseen data. Therefore, We introduce noise to bounding boxes, generating noisy boxes for training, thus enhancing model robustness on testing data. We propose a new paradigm to formulate the visual object tracking problem as a denoising learning process. However, tracking algorithms are usually asked to run in real-time, directly applying the diffusion model to object tracking would severely impair tracking speed. Therefore, we decompose the denoising learning process into every denoising block within a model, not by running the model multiple times, and thus we summarize the proposed paradigm as an in-model latent denoising learning process. Specifically, we propose a denoising Vision Transformer (ViT), which is composed of multiple denoising blocks. In the denoising block, template and search embeddings are projected into every denoising block as conditions. A denoising block is responsible for removing the noise in a predicted bounding box, and multiple stacked denoising blocks cooperate to accomplish the whole denoising process. Subsequently, we utilize image features and trajectory information to refine the denoised bounding box. Besides, we also utilize trajectory memory and visual memory to improve tracking stability. Experimental results validate the effectiveness of our approach, achieving competitive performance on several challenging datasets. The proposed in-model latent denoising tracker achieve real-time speed, rendering denoising learning

关键词：

来源：评论

学校读者我要写书评

暂无评论

Vision-Based Object Sorting in Dynamic Environments using YOLO for RoboCup ARM Challenge 2023 7

Vision-Based Object Sorting in Dynamic Environments using YO...

引用

7th IEEE Conference on Information and Communication Technology, CICT 2023

作者： Sahu, Yash Kumar Mittal, Radhika Patel, Deep Maiti, Chayan Sreekumar, M. Indian Institute of Information Technology Design and Manufacturing Centre for AI IoT and Robotics Department of Mechanical Engineering Kancheepuram Chennai India Indian Institute of Information Technology Design and Manufacturing Centre for AI IoT and Robotics Department of Computer Science Kancheepuram Chennai India

ISBN: (纸本)9798350305173

In the intersection of Artificial Intelligence (AI) and robotics, autonomous object manipulation has emerged as a pivotal research domain, finding applications in sectors as diverse as manufacturing and healthcare. While significant strides have been made in object identification, trajectory planning, and manipulation tasks, these advancements often fall short in real-world, dynamic settings, revealing a critical research gap in algorithmic adaptability and computational efficiency. This paper addresses the gap by applying YOLOv4, a state-of-the-art object detection AI algorithm, within the context of the RoboCup Autonomous Robot Manipulation (ARM) Challenge - a competitive platform designed to simulate real-world complexities in autonomous systems. The research objectives are twofold: firstly, the performance of YOLOv4 in object detection and classification under dynamically changing conditions is rigorously evaluated, employing varied training strategies in both simulated and real-world scenarios. Secondly, the robustness and adaptability of the algorithm in object manipulation tasks are scrutinized and facilitated by the real-world dynamics inherent to the competition. Empirical results demonstrate the algorithm's high accuracy, efficiency, and robustness in dynamic environments, substantiating its broader applicability in the development of autonomous systems. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Dating Wall Constructions from Brick Sizes in the Flooded Basement of a WW-II Submarine Bunker for the Digitization of Cultural Heritage

Dating Wall Constructions from Brick Sizes in the Flooded Ba...

引用

OCEANS

作者： Andreas Birk Frederike Buda Tim Hansen Computer Science and Electrical Engineering Constructor University Bremen Robotics

Results from the exploration of the flooded basement of the U-Boot Bunker Valentin are presented. The work is done in the context of the digitization of the bunker, which was constructed during World-War II using substantial amounts of forced labor and which is now a memorial. The presented research deals with walls in the basement for which it is of interest when they were constructed. Using a simple but efficient structured light system, the sizes of the bricks were measured. In combination with the changes in German standards for bricks after WW-II, this gives important clues about the time of construction of the investigated walls.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Drone-Based Predictive Structure Analysis: A Review

Drone-Based Predictive Structure Analysis: A Review

引用

2023 International Conference on Sustainable Emerging Innovations in Engineering and Technology, ICSEIET 2023

作者： Patil, Trushant Nandan, Durgesh Sharma, Kanhaiya Parashar, Deepak Department of Robotics and Automation Pune India Department of Electronics and Telecommunication Engg. Pune India Department of Computer Science & Engineering Pune India Faculty of Engineering Pune India

ISBN: (纸本)9798350329186

Drone-based predictive structure analysis is a rapidly growing field that has the potential to revolutionize infrastructure and building inspection. In this review of the research, we give a general overview of how drones could increase the effectiveness and precision of structural analysis, particularly for keeping an eye on difficult-to-reach locations. The usage of drones for 3D mapping, inspection, and monitoring of diverse constructions, such as concrete buildings, bridges, tunnels, and dams. The investigations show that drone-based inspection can produce 3D models and high-resolution photos, allowing for remote monitoring and the early identification of possible problems. The findings show that drones may greatly boost infrastructure management's safety and effectiveness, making them an important tool for predictive structural research. © 2023 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Optimizing Uncertainty Estimation on Scientific Visualizations Using Learning Models

Optimizing Uncertainty Estimation on Scientific Visualizatio...

引用

2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC Workshops 2024

作者： Almeida, Clara J. Guerrero-Pantoja, David Pautsch, Erik Rizzi, Silvio Thiruvathukal, George K. Pantoja, Maria University of California Santa Cruz Robotics and Electrical Engineering Santa Cruz United States University of California Santa Cruz Computer Science and Game Design Santa Cruz United States Loyola University Chicago Computer Science Chicago United States Argonne National Laboratory Argonne Leadership Computing Facility Lemont United States Computer Science and Software Engineering California Polytechnics State University San Luis Obispo United States

ISBN: (纸本)9798350355543

Scientific visualizations (SciVis) translate numerical and spatial data as images, enabling scientists to better understand the phenomena described by the data and gain insights that may be overlooked by statistical methods alone. In recent years, machine learning, particularly Deep Learning (DL), has driven significant advancements in SciVis. By combining classical numerical techniques with DL methods, we can create models that potentially achieve the accuracies needed for real-world industrial applications and simulations. However, DL models often exhibit undue confidence, particularly when faced with inputs that differ significantly from the data they were trained on, known as out-of-distribution (OOD) inputs. This can lead to misclassifications where the model still produces high confidence scores, reflected in elevated softmax output values, even though its prediction is incorrect. Ideally, in such cases, the model should indicate uncertainty, effectively saying, "I do not know."In this work, we enhance the trustworthiness of DL models by quantifying their uncertainty and enabling selective classification, where the model abstains from making predictions when uncertainty is high. Instead of providing a single prediction, our model outputs a prediction distribution, allowing users to determine whether human intervention or other actions are necessary for handling high-uncertainty outputs. This paper evaluates uncertainty quantification across three different tasks: predicting airfoil pressure and velocity using a Reynolds-Averaged Navier-Stokes (RANS) model on a custom airfoil dataset, image classification using the ImageNet1K dataset, and digit recognition using the MNIST handwritten digits dataset. © 2024 IEEE.

关键词： deep learning ensemble learning evidential learning scientific visualization uncertainty

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Lossless Compression Technique in Quantum Secure Direct Communication 1

A Comparative Study of Lossless Compression Technique in Qua...

引用

1st International Conference on Cyber Security and Computing, CyberComp 2024

作者： Ishak, Izlin Zuriani Wooi, Liew Seow Harun, Nur Ziadah Azahari, Nur Shahirah Ramli, Sofia Najwa Ghazali, Osman Politeknik Seberang Perai Department of Information Technology and Communication Pulau Pinang Malaysia Universiti Putra Malaysia Faculty of Computer Science and Information Technology Department of Computer Science Selangor Malaysia Raffles University Faculty of AI and Robotics Johor Malaysia Universiti Tun Hussein Onn Malaysia Faculty of Computer Science and Information Technology Department of Information Security and Web Technology Johor Malaysia School of Computing Universiti Utara Malaysia Kedah Malaysia

ISBN: (纸本)9798350387728

This is a comparative study to explore the performance of three lossless compression techniques, namely Run Length Encoding (RLE), Shannon-Fano Coding (SFC), and Huffman Coding (HC) in the context of a Hybrid M-Ary in Braided Single Stage (HMBSS) protocol communication scheme. The previous study on HMBSS using HC found that HC's compression technique has drawbacks, i.e., it is sensitive to errors and noise in the transmission channel and limited to certain data types. These drawbacks did affect the value of the compression rate limit. In this study, we evaluate the effectiveness of RLE, SFC, and HC in terms of data size reduction, coding efficiency, and data transmission time by comparing their results arising from the simulation experiments using Python coding in Jupiter Notebook and developed based on the HMBSS protocol algorithm. This simulation experiment starts with an initial six randomly generated datasets. Each dataset is tested using three compression techniques, which resulted in 18 different datasets. The results from the experiments reveal that SFC and HC consistently achieved a higher compression ratio and coding efficiency than RLE. These two compression techniques' algorithms effectively reduce the data size while preserving the original information. In our study, we also find that the data transmission time is increased with larger text sizes for all three compression techniques. This indicates that the larger the data size, the more transmission time is required with regard to what the compression technique's algorithm is. © 2024 IEEE.

关键词： Quantum communication

来源：评论

学校读者我要写书评

暂无评论

iDrone: IoT-Enabled Unmanned Aerial Vehicles for Detecting Wildfires Using Convolutional Neural Networks

引用

SN computer science 2022年第3期3卷 242页

作者： Sethuraman, Sibi Chakkaravarthy Tadkapally, Gaurav Reddy Mohanty, Saraju P. Subramanian, Anitha School of Computer Science and Engineering and Center of Excellence in Artificial Intelligence and Robotics (AIR) VIT-AP University Amaravati India Department of Computer Science and Engineering University of North Texas Denton 76207 TX United States School of Electronics Engineering and Center of Excellence in Artificial Intelligence and Robotics (AIR) VIT-AP University Amaravati India

The rise of global temperatures, over the past few decades, has disrupted the usual balance of nature. As a result of increasing temperatures, wildfires have destroyed millions of acres of land, thousands of structures, and homes. The pollution and toxic gases produced by the wildfires are carried out to thousands of miles, thus threatening the lives all around the world. Most wildfires occur due to anthropogenic factors, which cannot be predicted solely based on climate conditions. Henceforth, to detect wildfires before escalating, we propose iDrone, which is a wildfire detection system equipped with an end-to-end CNN image classification model: XtinguishNet, trained on a wildfire imagery dataset to detect the possible flames or smokes in an image. In addition, our approach also acquires the weather data and the intensity of the fire. Contrasting with existing wildfire detection systems, our proposed solution is a fusion of the Internet of Things (IoT) and Deep Learning, aiming to provide a one-stop solution for all the needs required to minimize the damage caused by wildfires. When validated and tested using various benchmark datasets, video surveillance, iDrone acquired a high accuracy of 98.36% with the least computational power. © 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： CNN computer vision Deep learning Forest fires iDrone Wildfires

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：