检索结果-内蒙古大学图书馆

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES 2024年第1期9卷 1524-1536页

作者： Yu, Zedong Wan, Weibing Ren, Maiyu Zheng, Xiuyuan Fang, Zhijun Shanghai Univ Engn Sci Sch Elect & Elect Engn Int Joint Lab Intelligent Sensing & Control Shanghai 201620 Peoples R China Donghua Univ Sch Comp Sci & Technol Shanghai 200051 Peoples R China

In the context of autonomous driving environment perception, multi-modal fusion plays a pivotal role in enhancing robustness, completeness, and accuracy, thereby extending the performance boundary of the perception system. However, directly applying LidAR-related algorithms to radar and camera fusion leads to significant challenges, such as radar sparsity, absence of height information, and noise, resulting in substantial performance loss. To address these issues, our proposed method, SparseFusion3d, utilizes a dual-branch feature-level fusion network that fully models sensor interactions, effectively mitigating the adverse effects of radar sparsity and noise on modality association. Additionally, to enhance modal correlations and accuracy while alleviating radar point cloud sparsity and measurement ambiguity, we introduce MSPCP, which compensates for point cloud offset. Moreover, we integrate Radar Painter to leverage image information and further enhance MSPCP. SparseFusion3d exhibits competitive performance compared to previous radar-camera fusion models, achieving approximately 1.5x inference speedup with similar performance to dense query methods, while also improving by 20.1% compared to the baseline approach.

关键词： Radar Three-dimensional displays Point cloud compression Radar detection Radar imaging Cameras object detection Multi-modal fusion 3d object detection intelligent driving vehicle radar camera

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Online Continual Learning for 3d object detection in Mobile Robotics

引用

JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS 2024年第4期110卷 1-16页

作者： Liu, Binhong Yao, dexin Yang, Rui Yan, Zhi Yang, Tao Northwestern Polytech Univ Unmanned Syst Res Inst Natl Key Lab Unmanned Aerial Vehicle Technol Integ Xian 710072 Peoples R China Univ Bourgogne Franche Comte CIAD UTBM UMR7533 F-90010 Belfort France

Continual learning addresses the challenge of acquiring and retaining knowledge over time across multiple tasks and environments. Previous research primarily focuses on offline settings where models learn through increasing tasks from samples paired with ground truth annotations. In this work, we focus on an unsolved, challenging, yet practical scenario, specifically, the semi-supervised online continual learning in autonomous driving and mobile robotics. In our settings, models are tasked with learning new distributions from streaming unlabeled samples and performing 3d object detection as soon as the LidAR point cloud arrives. Additionally, we conducted experiments on both the KITTI dataset, our newly built IUSL dataset and Canadian Adverse driving Conditions (CAdC) dataset. The results indicate that our method achieves a balance between rapid adaptation and knowledge retention, showcasing its effectiveness in the dynamic and complex environment of autonomous driving and mobile robotics. The developed ROS packages and IUSL dataset will be publicly available at: https://***/npu-ius-lab/OCL3d.

关键词： Semi-supervised Streaming learning Online continual learning 3d object detection

来源：评论

学校读者我要写书评

暂无评论

RCF-TP: Radar-Camera Fusion With Temporal Priors for 3d object detection

引用

IEEE ACCESS 2024年 12卷 127212-127223页

作者： Miron, Yakov drews, Florian Faion, Florian di Castro, dotan Klein, Itzik Bosch Res IL-3109701 Haifa Israel Univ Haifa Hatter Dept Marine Technol IL-3109701 Haifa Israel Bosch Res D-71272 Renningen Germany

Sensor fusion is an important method for achieving robust perception systems in autonomous driving, Internet of things, and robotics. Most multi-modal 3d detection models assume the data is synchronized between the sensors and do not necessarily have real-time capabilities. We propose RCF-TP, an asynchronous, modular, real-time multi-modal architecture, to fuse cameras and radars for 3d object detection, with sensor fault mitigation and extreme weather conditions handling. Our dedicated feature extractors can be trained assuming either a regular or an irregular bird's-eye-view grid or with different grid resolutions, such that the fusion module is agnostic to both. These extracted features are correlated to the other modality features or to another sensor of the same modality, and eventually a detection head that exploits rich multi-modal features could be applied at any time to produce bounding box predictions. Experimental results show the effectiveness of our fusion module. It improves detection performance for higher radar grid resolution, can operate under sensor faults without performance degradation, and improves pedestrian detection when our dataset combination strategy is implemented during training.

关键词： Radar Feature extraction Three-dimensional displays Cameras Radar imaging Radar detection Sensor phenomena and characterization Sensor fusion Self-supervised learning 3d object detection self-supervised learning sensor dropout

来源：评论

学校读者我要写书评

暂无评论

MCHFormer: A Multi-Cross Hybrid Former of Point-Image for 3d object detection

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES

引用

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES 2024年第1期9卷 383-394页

作者： Cao, Feng Xue, Jun Tao, Chongben Luo, Xizhao Gao, Zhen Zhang, Zufeng Zheng, Sifa Zhu, Yuan Shanxi Univ Sch Comp & Informat Technol Taiyuan 030006 Peoples R China Suzhou Univ Sci & Technol Sch Elect & Informat Engn Suzhou 215009 Peoples R China Tsinghua Univ Suzhou Automobile Res Inst Suzhou 215134 Peoples R China Soochow Univ Sch Comp Sci & Technol Suzhou 215006 Peoples R China McMaster Univ Fac Engn Hamilton ON Canada Tsinghua Univ Beijing 100084 Peoples R China Tongji Univ Coll Automot Studies Shanghai 201804 Peoples R China

Mismatch often occurs between local and global information in multimodal data during downscaling transformation, which results in the loss of localization information. A Multi-Cross Hybrid Former (MCHFormer) of point-image is proposed for 3d object detection in autonomous driving, which cross-fuses LidAR with cameras at multiple levels. Specifically, the voxelized point cloud is firstly extracted through a dual-Stream Feature Extraction (dSFE) network. Local fine-grained area information is integrated into the global feature information, which results in a multi-layered Bird's Eye View (BEV). Meanwhile, the raw coordinates of points are incorporated into point-wise features through position coding. Then, point features are projected onto image and BEV features to obtain highly coupled multimodal information, which achieves alignment of point cloud with image information. Finally, a multi-cross Transformer fuses multiple unimodal data into a hybrid representation with more spatial awareness, which achieves accurate 3d object detection. MCHFormer are conducted extensive comparative experiments with other State-Of-The-Art (SOTA) algorithms on the KITTI, NuScenes, Waymo datasets and real road scenes. Experimental results show that the proposed algorithm not only has better accuracy and generalization capability, but also has accurate detection effect on real road scenarios.

关键词： Three-dimensional displays Feature extraction Transformers object detection Point cloud compression Laser radar detectors 3d object detection automatic driving multimodal Fusion multi-Cross Feature Extraction Transformer

来源：评论

学校读者我要写书评

暂无评论

Multi-Modal 3d object detection by Box Matching

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2024年第12期25卷 19917-19928页

作者： Liu, Zhe Ye, Xiaoqing Zou, Zhikang He, Xinwei Tan, Xiao ding, Errui Wang, Jingdong Bai, Xiang Huazhong Univ Sci & Technol Sch Elect Informat & Commun Wuhan 430074 Peoples R China Baidu Inc Beijing 100085 Peoples R China Huazhong Agr Univ Coll Informat Wuhan 430070 Peoples R China Huazhong Univ Sci & Technol Sch Software Wuhan 430074 Peoples R China

Multi-modal 3d object detection has received growing attention as the information from different sensors like LidAR and cameras are complementary. Most fusion methods for 3d detection rely on an accurate alignment and calibration between 3d point clouds and RGB images. However, such an assumption is not reliable in a real-world self-driving system, as the alignment between different modalities is easily affected by asynchronous sensors and disturbed sensor placement. We propose a novel Fusion network by Box Matching (FBMNet) for multi-modal 3d detection, which provides an alternative way for cross-modal feature alignment by learning the correspondence at the bounding box level to free up the dependency of calibration during inference. With the learned assignments between 3d and 2d object proposals, the fusion for detection can be effectively performed by combining their ROI features. Extensive experiments on the nuScenes dataset demonstrate that our method is much more robust in dealing with challenging cases such as asynchronous sensors, misaligned sensor placement, and degenerated camera images than existing fusion methods. We hope that our FBMNet could provide an available solution to dealing with these challenging cases for safety in real autonomous driving scenarios.

关键词： Multi-modal 3d object detection feature alignment box matching

来源：评论

学校读者我要写书评

暂无评论

PSNS-SSd: Pixel-Level Suppressed Nonsalient Semantic and Multicoupled Channel Enhancement Attention for 3d object detection

引用

IEEE ROBOTICS ANd AUTOMATION LETTERS 2024年第1期9卷 603-610页

作者： Song, Xiaogang Zhou, Zhenhua Zhang, Lei Lu, Xiaofeng Hei, Xinhong Xian Univ Technol Sch Comp Sci & Engn Xian 710048 Peoples R China Univ Shaanxi Prov Engn Res Ctr Human Machine Integrat Intelligent Ro Xian 710048 Peoples R China Aviation Ind Corp China Xian Aeronaut Comp Tech Res Inst Xian 710048 Peoples R China

In the field of 3d object detection, voxel-based methods are the most commonly used and exhibit high accuracy. However, point-based networks, which have the capability to preserve the original point features, are unable to surpass the accuracy achieved by voxel-based methods. We observed that during downsampling, too many useless background points are retained in the prevailing farthest point sample, which interferes with the accuracy of the point-based network and introduces redundant computation. To address this issue, in this letter, a point-based single-stage 3d detection network called PSNS-SSd is designed, and its accuracy rivals the accuracy of the two-stage voxel-based methods. Our method leverages the semantic information that suppresses nonsalient features used for guidance and discards background points during the downsampling process. In addition, we propose two modules, namely, linear-conv channel-enhanced self-attention for point cloud feature extraction and global-local coupling attention for center prediction, to enhance the performance and efficiency of our model. Our PSNS-SSd model is extensively tested on the most widely available 3d dataset-the KITTI dataset. The experimental results validate the three proposed modules in terms of accuracy for small object detection and speed, and the model achieves accuracy levels that are on par with the accuracy levels of cutting-edge voxel-based techniques. Moreover, our detector can achieve a speed of 70 FPS on the KITTI dataset, making it a promising candidate for practical applications.

关键词： Feature extraction Three-dimensional displays Semantics Point cloud compression object detection detectors Proposals 3d object detection high efficiency nonsalient semantic point-based point clouds

来源：评论

学校读者我要写书评

暂无评论

VSL-Net: Voxel structure learning for 3d object detection

引用

AdVANCEd ENGINEERING INFORMATICS 2024年 59卷

作者： Cao, Feng Zhou, Feng Tao, Chongben Xue, Jun Gao, Zhen Zhang, Zufeng Zhu, Yuan Shanxi Univ Sch Comp & Informat Technol Taiyuan 030006 Peoples R China Suzhou Univ Sci & Technol Sch Elect & Informat Engn Suzhou 215009 Peoples R China Tsinghua Univ Suzhou Automobile Res Inst Suzhou 215134 Peoples R China McMaster Univ Fac Engn Hamilton ON L8S 0A Canada Tsinghua Univ Dept Automat Beijing 100084 Peoples R China Tongji Univ Coll Automot Studies Shanghai 201804 Peoples R China

Current detection methods with single stage generally lack contextual structure information, the classification and location confidence are inconsistent, which are not able to achieve accurate dynamic multi -object detection. Therefore, a VSL-Net method is proposed based on single -stage detection framework, which maps voxels to the point cloud space through voxel structure learning for foreground point segmentation and center estimation. In the voxel feature extraction network, the submanifold sparse convolution is proposed to improve convolution efficiency by reducing operation of null region. In the detection network, a position alignment module is proposed to sample and interpolate in Bird's Eye View. Characteristics of surrounding points are integrated to further improve the accuracy of detection. The proposed VSL-Net method was compared with other excellent method on the Kitti, Waymo and Nuscene datasets. The vehicle detection accuracy of VSL-Net reached 92.12%. Ablation and accuracy detection experiments have completed on a real physical platform. Experiment results showed that the proposed method had portability, strong generalization ability and high detection accuracy.

关键词： 3d object detection Single stage algorithm Voxel structure learning Submanifold sparse convolution

来源：评论

学校读者我要写书评

暂无评论

Transformer-Based Stereo-Aware 3d object detection From Binocular Images

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2024年第12期25卷 19675-19687页

作者： Sun, Hanqing Pang, Yanwei Cao, Jiale Xie, Jin Li, Xuelong Chinese Acad Sci Changchun Inst Opt Fine Mech & Phys Changchun 130033 Peoples R China Tianjin Univ Sch Elect & Informat Engn Tianjin Key Lab Brain Inspired Intelligence Techn Tianjin 300072 Peoples R China Chongqing Univ Sch Big Data & Software Engn Chongqing 400050 Peoples R China Northwestern Polytech Univ Sch Comp Sci Xian 710060 Peoples R China Northwestern Polytech Univ Sch Artificial Intelligence Opt & Elect iOPEN Xian 710060 Peoples R China

Transformers have shown promising progress in various visual object detection tasks, including monocular 2d/3d detection and surround-view 3d detection. More importantly, the attention mechanism in the Transformer model and the 3d information extraction in binocular stereo are both similarity-based. However, directly applying existing Transformer-based detectors to binocular stereo 3d object detection leads to slow convergence and significant precision drops. We argue that a key cause of that defect is that existing Transformers ignore the binocular-stereo-specific image correspondence information. In this paper, we explore the model design of Transformers in binocular 3d object detection, focusing particularly on extracting and encoding task-specific image correspondence information. To achieve this goal, we present TS3d, a Transformer-based Stereo-aware 3d object detector. In the TS3d, a disparity-Aware Positional Encoding (dAPE) module is proposed to embed the image correspondence information into stereo features. The correspondence is encoded as normalized sub-pixel-level disparity and is used in conjunction with sinusoidal 2d positional encoding to provide the 3d location information of the scene. To enrich multi-scale stereo features, we propose a Stereo Preserving Feature Pyramid Network (SPFPN). The SPFPN is designed to preserve the correspondence information while fusing intra-scale and aggregating cross-scale stereo features. Our proposed TS3d achieves a 41.29% Moderate Car detection average precision on the KITTI test set and takes 88 ms to detect objects from each binocular image pair. It is competitive with advanced counterparts in terms of both precision and inference speed.

关键词： Stereo vision 3d object detection Transformer positional encoding feature pyramid image correspondence

来源：评论

学校读者我要写书评

暂无评论

CL-fusionBEV: 3d object detection method with camera-LidAR fusion in Bird's Eye View

引用

COMPLEX & INTELLIGENT SYSTEMS 2024年第6期10卷 7681-7696页

作者： Shi, Peicheng Liu, Zhiqiang dong, Xinlong Yang, Aixi Anhui Polytech Univ Sch Mech & Automot Engn Wuhu 241000 Peoples R China Nanjing Univ Aeronaut & Astronaut Coll Mech & Elect Engn Nanjing 210016 Peoples R China Zhejiang Univ Polytech Inst Hangzhou 310058 Peoples R China

In the wave of research on autonomous driving, 3d object detection from the Bird's Eye View (BEV) perspective has emerged as a pivotal area of focus. The essence of this challenge is the effective fusion of camera and LidAR data into the BEV. Current approaches predominantly train and predict within the front view and Cartesian coordinate system, often overlooking the inherent structural and operational differences between cameras and LidAR sensors. This paper introduces CL-FusionBEV, an innovative 3d object detection methodology tailored for sensor data fusion in the BEV perspective. Our approach initiates with a view transformation, facilitated by an implicit learning module that transitions the camera's perspective to the BEV space, thereby aligning the prediction module. Subsequently, to achieve modal fusion within the BEV framework, we employ voxelization to convert the LidAR point cloud into BEV space, thereby generating LidAR BEV spatial features. Moreover, to integrate the BEV spatial features from both camera and LidAR, we have developed a multi-modal cross-attention mechanism and an implicit multi-modal fusion network, designed to enhance the synergy and application of dual-modal data. To counteract potential deficiencies in global reasoning and feature interaction arising from multi-modal cross-attention, we propose a BEV self-attention mechanism that facilitates comprehensive global feature operations. Our methodology has undergone rigorous evaluation on a substantial dataset within the autonomous driving domain, the nuScenes dataset. The outcomes demonstrate that our method achieves a mean Average Precision (mAP) of 73.3% and a nuScenes detection Score (NdS) of 75.5%, particularly excelling in the detection of cars and pedestrians with high accuracies of 89% and 90.7%, respectively. Additionally, CL-FusionBEV exhibits superior performance in identifying occluded and distant objects, surpassing existing comparative methods.

关键词： Bird's Eye View (BEV) perception 3d object detection Attention mechanism Autonomous driving

来源：评论

学校读者我要写书评

暂无评论

GraphAlign plus plus : An Accurate Feature Alignment by Graph Matching for Multi-Modal 3d object detection

引用

IEEE TRANSACTIONS ON CIRCUITS ANd SYSTEMS FOR VIdEO TECHNOLOGY 2024年第4期34卷 2619-2632页

作者： Song, Ziying Jia, Caiyan Yang, Lei Wei, Haiyue Liu, Lin Beijing Jiaotong Univ Sch Comp & Informat Technol Beijing Key Lab Traff Data Anal & Min Beijing 100044 Peoples R China Tsinghua Univ State Key Lab Automot Safety & Energy Beijing 050018 Peoples R China Hebei Univ Sci & Technol Sch Informat Sci & Engn Shijiazhuang 050018 Peoples R China

LidAR and camera are complementary sensors for 3d object detection in autonomous driving. However, it is challenging to explore the unnatural interaction between point clouds and images, and the critical factor is how to conduct feature alignment of these heterogeneous modalities. Currently, many methods achieve feature alignment through projection calibration, without accounting for the impact of sensors misalignment errors, resulting in sub-optimal performance. In this paper, we present GraphAlign++, a more accurate feature alignment framework for 3d object detection by graph matching. Specifically, we construct the nearest neighbor relationship by calculating Euclidean distances of point cloud features within the subspaces. Through the projection calibration between the image and point cloud pairs, we project the nearest neighbors of point cloud features onto the corresponding image. Then by matching the nearest neighbors of a single point-feature of the point cloud with multiple pixel-features of the image, we search for a more appropriate feature alignment. In addition, we provide a self-attention module to enhance the weights of significant relations to fine-tune the feature alignment between these two heterogeneous modalities. Extensive experiments on nuScenes benchmark demonstrate the effectiveness and efficiency of GraphAlign++. Notably, due to the more accurate feature alignment, which contributes to increase mAP by 3.10% on KITTI test hard level, our method is remarkably beneficial for long-range object detection.

关键词： Three-dimensional displays Point cloud compression Feature extraction object detection Laser radar Semantics Cameras 3d object detection multi-modal fusion feature alignment local graphs

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：