检索结果-内蒙古大学图书馆

Thermal Chameleon: Task-Adaptive Tone-Mapping for Radiometric Thermal-Infrared Images

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第12期9卷 10827-10834页

作者： Lee, Dong-Guw Kim, Jeongyun Cho, Younggun Kim, Ayoung SNU Dept Mech Engn Seoul 08826 South Korea Inha Univ Dept Elect Engn Incheon 22212 South Korea

Thermal Infrared (TIR) imaging provides robust perception for navigating in challenging outdoor environments but faces issues with poor texture and low image contrast due to its 14/16-bit format. Conventional methods utilize various tone-mapping methods to enhance contrast and photometric consistency of TIR images, however, the choice of tone-mapping is largely dependent on knowing the task and temperature dependent priors to work well. In this paper, we present Thermal Chameleon (TCNet), a task-adaptive tone-mapping approach for RAW 14-bit TIR images. Given the same image, TCNet tone-maps different representations of TIR images tailored for each specific task, eliminating the heuristic image rescaling preprocessing and reliance on the extensive prior knowledge of the scene temperature or task-specific characteristics. TCNet exhibits improved generalization performance across object detection and monocular depth estimation, with minimal computational overhead and modular integration to existing architectures for various tasks.

关键词： Image coding Radiometry Temperature distribution Cameras Feature extraction Adaptive systems Training Thermal noise Sensor phenomena and characterization Object detection Deep learning for visual perception object detection representation learning segmentation and categorization

来源：评论

学校读者我要写书评

暂无评论

Deforming Garment Classification With Shallow Temporal Extraction and Tree-Based Fusion

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第2期9卷 1114-1121页

作者： Huang, Li Yang, Tong Jiang, Rongxin Tian, Xiang Zhou, Fan Chen, Yaowu Zhejiang Univ Inst Adv Digital Technol & Instrumentat Hangzhou 310027 Peoples R China Zhejiang Univ State Key Lab Ind Control & Technol Hangzhou 310027 Peoples R China Minist Educ China Embedded Syst Engn Res Ctr Hangzhou 310027 Peoples R China Zhejiang Prov Key Lab Network Multimedia Technol Hangzhou 310027 Peoples R China

A novel RGB-based continuous perception garment classification approach is proposed in this letter, with the aim of identifying the correct category of the garment from a set of categories. It has been observed that treating a video of the continuous deformation of cloth as a set of disordered static figures leads to low classification precision. On the contrary, investigating the temporal information between frames can significantly improve the quality of extracted features and increase classification performance. In this regard, we propose a hybrid temporal fusion RGB-based algorithm, including an improved image-level shallow temporal feature extraction module (STEM) and a binary-tree fusion module (BiTF) for adaptive feature fusion. STEM incorporates multi-scale optical flow and long-short-term memorised information to capture both static features in every single image and dynamic features in consecutive images. BiTF constructs a tree-shaped structure to fuse an arbitrary number of extracted features in a video. The proposed algorithm achieves 97.5% accuracy on the public dataset, surpassing the state-of-the-art RGB-based classification algorithm by 44.0% and achieving comparable performance with RGBD-based algorithms.

关键词： Object detection segmentation and categorization deep learning for visual perception shallow temporal feature extraction binary-tree fusion

来源：评论

学校读者我要写书评

暂无评论

Improving Self-Supervised Learning of Transparent Category Poses With Language Guidance and Implicit Physical Constraints

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第9期9卷 8114-8121页

作者： Wang, Pengyuan Garattoni, Lorenzo Meier, Sven Navab, Nassir Busam, Benjamin Tech Univ Munich Chair Comp Aided Med Procedures D-80804 Munich Germany Toyota Motor Europe B-1930 Brussels Belgium

Accurate object pose estimation is crucial for robotic applications and recent trends in category-level pose estimation show great potential for applications encountering a large variety of similar objects, often encountered in home environments. While common in such environments, photometrically challenging objects with transparency such as glasses are poorly handled by current methods. Especially using self-supervision to bridge the sim2real domain gap is difficult for transparent objects due to strong background changes and depth artifacts. To address this, we propose a novel pipeline which takes language guidance and implicit physical constraints for 2D and 3D self-supervisions. In specific, we utilize language guidance to obtain accurate 2D object segmentation which is robust to background changes. Further 3D self-supervisions are achieved by contact constraint and normal constraint from polarization inputs with a differentiable renderer. Instead of explicitly leveraging the depth measurements, we reason about implicit physical constraints for self-supervisions. Extensive experiments superior performance of our self-supervision approach over baselines on both the self-collected dataset and public benchmarks, addressing photometric challenges.

关键词： Object detection segmentation and categorization RGB-D perception perception for grasping and manipulation Object detection segmentation and categorization RGB-D perception perception for grasping and manipulation

来源：评论

学校读者我要写书评

暂无评论

Sphere-Graph: A Compact 3D Topological Map for Robotic Navigation and segmentation of Complex Environments

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第3期9卷 2567-2574页

作者： Spencer, Meryl Sawtell, Reid Kitchen, Sarah MichiganTech Univ Michigan Tech Res Inst Ann Arbor MI 48104 USA

Topological maps are a common framework for enabling autonomous robotic navigation. To be effective for robotic exploration the maps must be able to be generated quickly and compact enough to store on lightweight hardware. Here we propose a novel 3D topological map called Sphere-Graph which has adaptive edge lengths, can be quickly generated, and can be used to semantically identify hallways and rooms to produce a compact representation of complex environments. We give examples of the Sphere-Graph representation of large 3D urban and cave environments.

关键词： Three-dimensional displays Robots Navigation Topology Semantics Task analysis Semantic segmentation Autonomous vehicle navigation mapping object detection segmentation and categorization

来源：评论

学校读者我要写书评

暂无评论

UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection With Sparse LiDAR and Large Domain Gaps

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第12期9卷 11210-11217页

作者： Wozniak, Maciej K. Hansson, Mattias Thiel, Marko Jensfelt, Patric KTH Royal Inst Technol Div Robot Percept & Learning S-11428 Stockholm Sweden Hamburg Univ Technol Inst Tech Logist D-21073 Hamburg Germany

In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains.

关键词： Feature extraction Three-dimensional displays Laser radar Object detection Robot sensing systems Adaptation models Point cloud compression Solid modeling Training Mobile robots Deep learning for visual perception object detection segmentation and categorization

来源：评论

学校读者我要写书评

暂无评论

A Sim-to-Real Instance segmentation Framework for Densely Stacked Cartons

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第6期9卷 5775-5782页

作者： Wang, Xu Li, Junliang Zheng, Wenli Luo, Huanliang Wang, Faping Chen, Yongquan Huang, Rui Chinese Univ Hong Kong Shenzhen Inst Artificial Intelligence & Robot Soc Shenzhen 518172 Peoples R China Dapeng Customs Peoples Republ China Shenzhen 518083 Peoples R China Shenzhen Haixing Intelligent Driving Technol Co Lt Shenzhen 518110 Guangdong Peoples R China

Robotic picking systems in automated logistics require accurate segmentation and localization of densely stacked cartons. However, the lack of comprehensive and diverse datasets for this task poses a significant challenge. Furthermore, existing instance segmentation methods struggle to meet the accuracy requirements due to the unique characteristics of cartons, including their large quantity, small spatial gaps, similar postures, and complex textures. To address these issues, this letter proposes a sim-to-real instance segmentation framework specifically designed for densely stacked cartons. Synthetic data of densely stacked cartons is generated using an automated simulation pipeline, resulting in the Densely Stacked Object Dataset (DSOD) with 40 000 data samples and 1 209 951 individually segmented cartons in various environmental settings. The instance segmentation network employs gated fully fusion and disentangled non-local modules to prioritize salient boundaries, improving segmentation accuracy. Experimental evaluation on both the DSOD and an existing real-world carton dataset demonstrates that our method outperforms state-of-the-art techniques, achieving an impressive performance of 85.2 on AP50 and 78.7 on AP75. These findings highlight the effectiveness of our sim-to-real framework for accurately segmenting and localizing densely stacked cartons.

关键词： Instance segmentation Pipelines Synthetic data Location awareness Grasping Task analysis Stacking Object detection segmentation and categorization deep learning for visual perception data sets for robotic vision

来源：评论

学校读者我要写书评

暂无评论

MovingCables: Moving Cable segmentation Method and Dataset

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第8期9卷 6991-6998页

作者： Holesovsky, Ondrej Skoviera, Radoslav Hlavac, Vaclav Czech Tech Univ Czech Inst Informat Robot & Cybernet Prague 16000 Czech Republic Czech Tech Univ Fac Elect Engn Prague 16627 Czech Republic

Manipulating cluttered cables, hoses or ropes is challenging for both robots and humans. Humans often simplify these perceptually challenging tasks by pulling or pushing tangled cables and observing the resulting motions. We propose to use a similar interactive perception principle to aid robotic cable manipulation. A fundamental building block of such an endeavor is a cable motion segmentation method that densely labels moving cable image pixels. This letter presents MovingCables, a moving cable dataset, which we hope will motivate the development and evaluation of cable motion segmentation algorithms. The dataset consists of real-world image sequences automatically annotated with ground truth segmentation masks and optical flow. In addition, we propose a cable motion segmentation method and evaluate its performance on the new dataset.

关键词： Motion segmentation Cables Image segmentation Optical flow Robots Hoses Computer vision Data sets for robotic vision deep learning for visual perception object detection segmentation and categorization cable motion optical flow

来源：评论

学校读者我要写书评

暂无评论

USD-SLAM: A Universal Visual SLAM Based on Large segmentation Model in Dynamic Environments

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第12期9卷 11810-11817页

作者： Wang, Jingwei Ren, Yizhang Li, Zhiwei Xie, Xiaoming Chen, Zilong Shen, Tianyu Liu, Huaping Wang, Kunfeng Beijing Univ Chem Technol Coll Informat Sci & Technol Beijing 100029 Peoples R China Beihang Univ Coll Instrumental Sci & Optoelect Engn Beijing 100191 Peoples R China Tsinghua Univ Dept Comp Sci & Technol Beijing 100084 Peoples R China

Visual Simultaneous Localization and Mapping (SLAM) has been widely adopted in autonomous driving and robotics. While most SLAM systems operate effectively in static or low-dynamic environments, achieving precise pose estimation in diverse unknown dynamic environments continues to pose a significant challenge. This letter introduces an advanced universal visual SLAM system (USD-SLAM) that combines a universal large segmentation model with a 3D spatial motion state constraint module to accurately handle any dynamic objects present in the environment. Our system first employs a large segmentation model guided by precise prompts to identify movable regions accurately. Based on the identified movable object regions, 3D spatial motion state constraints are exploited to remove the moving object regions. Finally, the moving object regions are excluded for subsequent tracking, localization, and mapping, ensuring stable and high-precision pose estimation. Experimental results demonstrate that our method can robustly operate in various dynamic and static environments without additional training, providing higher localization accuracy compared to other advanced dynamic SLAM systems.

关键词： Image segmentation Simultaneous localization and mapping Dynamics Three-dimensional displays Motion segmentation Semantics Feature extraction Vehicle dynamics Location awareness Cameras SLAM localization object detection segmentation and categorization

来源：评论

学校读者我要写书评

暂无评论

Radar Instance Transformer: Reliable Moving Instance segmentation in Sparse Radar Point Clouds

引用

IEEE TRANSACTIONS ON ROBOTICS 2024年 40卷 2357-2372页

作者： Zeller, Matthias Sandhu, Vardeep S. Mersch, Benedikt Behley, Jens Heidingsfeld, Michael Stachniss, Cyrill CARIAD SE D-53115 Bonn Germany Univ Bonn D-53115 Bonn Germany CARIAD SE D-71297 Monsheim Germany Univ Oxford Dept Engn Sci Oxford OX1 2JD England Lamarr Inst Machine Learning & Artificial Intellig Dortmund Germany

The perception of moving objects is crucial for autonomous robots performing collision avoidance in dynamic environments. LiDARs and cameras tremendously enhance scene interpretation but do not provide direct motion information and face limitations under adverse weather. Radar sensors overcome these limitations and provide Doppler velocities, delivering direct information on dynamic objects. In this article, we address the problem of moving instance segmentation in radar point clouds to enhance scene interpretation for safety-critical tasks. Our radar instance transformer enriches the current radar scan with temporal information without passing aggregated scans through a neural network. We propose a full-resolution backbone to prevent information loss in sparse point cloud processing. Our instance transformer head incorporates essential information to enhance segmentation but also enables reliable, class-agnostic instance assignments. In sum, our approach shows superior performance on the new moving instance segmentation benchmarks, including diverse environments, and provides model-agnostic modules to enhance scene interpretation.

关键词： Radar Point cloud compression Feature extraction Transformers Task analysis Encoding Sensors Deep learning in robotics and automation object detection radar perception segmentation and categorization semantic scene understanding

来源：评论

学校读者我要写书评

暂无评论

Prior Information-Assisted Neural Network for Point Cloud segmentation in Human-Robot Interaction Scenarios

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2024年第5期9卷 4186-4193页

作者： Lin, Jingxin Zhong, Kaifan Gong, Tao Zhang, Xianmin Wang, Nianfeng South China Univ Technol Guangdong Prov Key Lab Precis Equipment & Mfg Tech Guangzhou 510640 Peoples R China Shenzhen Polytech Shenzhen Peoples R China

This letter proposes a prior information-assisted (PIA) point cloud segmentation network that can be effectively applied to point cloud segmentation applications in human-robot interaction scenario. The joint angles of the robots are used as prior information, which is fed into the network as an additional input in the form of a vector along with the actual point cloud of the target scene. The PIA network incorporates a point cloud generation network and a simulation-assisted segmentation network. Based on the joint angles of the robots, the generation network generates a point cloud for a simulated scene without people or obstacles. Several new loss functions are proposed to train the point cloud generation network. The simulation-assisted segmentation network extracts and compares the features of the actual point cloud with those of the simulated point cloud and segments the actual point cloud. Experiments conducted on a homemade point cloud dataset involving an industrial scene demonstrate that the proposed approach can achieve significantly improved segmentation performance and network robustness.

关键词： AI-Based methods object detection segmentation and categorization safety in HRI

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：