检索结果-内蒙古大学图书馆

E2EK: End-to-End Regression Network Based on Keypoint for 6D Pose Estimation

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第3期7卷 6526-6533页

作者： Lin, Shifeng Wang, Zunran Ling, Yonggen Tao, Yidan Yang, Chenguang South China Univ Technol Sch Automat Sci & Engn Guangzhou 510000 Guangdong Peoples R China Tencent Robot X Shenzhen 518054 Peoples R China Shanghai Jiao Tong Univ Shanghai 200240 Peoples R China City Univ Hong Kong Hong Kong Peoples R China Univ West England Bristol Robot Lab Bristol BS16 1QY Avon England

The methods based on deep learning are the mainstream of 6D object pose estimation, which mainly include direct regression and two-stage pipelines. The former are keen by many scholarsat first due to their simplicity and differentiability to poses, but they usually lack in accuracy when compared with the latter that estimate the intermediate variables relating to geometries such as object keypoints or 2D-3D correspondence before PnP/RANSAC algorithm. However, the loss function of the two-stage method is non-differentiable to the 6D pose, which is hard to apply in the tasks requiring the differentiable poses. To overcome the disadvantages of the above methods, we propose an end-to-end regression network based on keypoints for 6D pose estimation. Specifically, we supervise the point-wise key point offsets that help the network to learn the geometric information and directly regress the 6D pose through aggregating keypoints to achieve differentiability to the pose. Furthermore, we improve the sampling method by sampling points around objects that benefits the small object and design a unit loss function that helps the learning of the keypoints. Experimental results show that our approach outperforms most methods on LM, LM-O and YCB-V datasets.

关键词： deep learning for visual perception pose estimation RGB-D perception

来源：评论

学校读者我要写书评

暂无评论

Adaptive Cost Volume Fusion Network for Multi-Modal Depth Estimation in Changing Environments

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第2期7卷 5095-5102页

作者： Park, Jinsun Jeong, Yongseop Joo, Kyungdon Cho, Donghyeon Kweon, In So Pusan Natl Univ PNU Sch Comp Sci & Engn Busan 46241 South Korea Korea Adv Inst Sci & Technol Robot Program Daejeon 34141 South Korea UNIST Artificial Intelligence Grad Sch Ulsan 44919 South Korea UNIST Dept Comp Sci & Engn Ulsan 44919 South Korea Chungnam Natl Univ CNU Dept Elect Engn Daejeon 34134 South Korea Korea Adv Inst Sci & Technol Sch Elect Engn Daejeon 34141 South Korea

In this letter, we propose an adaptive cost volume fusion algorithm for multi-modal depth estimation in changing environments. Our method takes measurements from multi-modal sensors to exploit their complementary characteristics and generates depth cues from each modality in the form of adaptive cost volumes using deep neural networks. The proposed adaptive cost volume considers sensor configurations and computational costs to resolve an imbalanced and redundant depth bases problem of conventional cost volumes. We further extend its role to a generalized depth representation and propose a geometry-aware cost fusion algorithm. Our unified and geometrically consistent depth representation leads to an accurate and efficient multi-modal sensor fusion, which is crucial for robustness to changing environments. To validate the proposed framework, we introduce a new multi-modal depth in changing environments (MMDCE) dataset. The dataset was collected by our own vehicular system with RGB, NIR, and LiDAR sensors in changing environments. Experimental results demonstrate that our method is robust, accurate, and reliable in changing environments. Our codes and dataset are available at our project page.(1)

关键词： AI-Based methods data sets for robotic vision deep learning for visual perception

来源：评论

学校读者我要写书评

暂无评论

SCVP: learning One-Shot View Planning via Set Covering for Unknown Object Reconstruction

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第2期7卷 1463-1470页

作者： Pan, Sicong Hu, Hao Wei, Hui Fudan Univ Sch Comp Sci Shanghai Key Lab Data Sci Lab Algorithms Cognit Models Shanghai 200438 Peoples R China

The view planning (VP) problem in robotic active vision enables a robot system to automatically perform object reconstruction tasks. Lacking prior knowledge, next-best-view (NBV) methods are typically used to plan a view sequence, with the goal of covering as many object surface areas as possible in an unknown environment. However, such methods have two problems: (1) they are unable to perform global path planning;and (2) the reconstruction process is inefficient because of time-consuming ray casting and high movement cost. We propose a neural network, SCVP, to pre-learn prior knowledge via set covering (SC) based training so as to achieve one-shot view planning. The SCVP network takes the volumetric occupancy grid as input and directly predicts a small (ideally minimum) number of views that cover all surface areas. Given object 3D models as a priori geometric knowledge, the training dataset is automatically labeled by the set covering optimization method. We propose a global path planning method to reconstruct objects without redundant movement. Comparative experiments on multiple datasets of 3D models show that, under the condition of similar or better surface coverage, the proposed method can outperform state-of-the-art NBV methods in terms of movement cost and inference time. Real-world experiments confirm that the proposed method can achieve faster object reconstruction than other methods.

关键词： Autonomous agents deep learning for visual perception motion and path planning data sets for robotic vision computer vision for automation

来源：评论

学校读者我要写书评

暂无评论

BIMS-PU: Bi-Directional and Multi-Scale Point Cloud Upsampling

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第3期7卷 7447-7454页

作者： Bai, Yechao Wang, Xiaogang Ang Jr, Marcelo H. Rus, Daniela Natl Univ Singapore Singapore 119077 Singapore MIT 77 Massachusetts Ave Cambridge MA 02139 USA

The learning and aggregation of multi-scale features are essential in empowering neural networks to capture the fine-grained geometric details in the point cloud upsampling task. Most existing approaches extract multi-scale features from a point cloud of a fixed resolution, hence obtain only a limited level of details. Though an existing approach aggregates a feature hierarchy of different resolutions from a cascade of upsampling sub-network, the training is complex with expensive computation. To address these issues, we construct a new point cloud upsampling pipeline called RIMS-PU that integrates the feature pyramid architecture with a bi-directional up and downsampling path. Specifically, we decompose the up/downsampling procedure into several up/downsampling sub-steps by breaking the target sampling factor into smaller factors. The multi-scale features are naturally produced in a parallel manner and aggregated using a fast feature fusion method. Supervision signal is simultaneously applied to all upsampled point clouds of different scales. Moreover, we formulate a residual block to ease the training of our model. Extensive quantitative and qualitative experiments on different datasets show that our method achieves superior results to state-of-the-art approaches. Last but not least, we demonstrate that point cloud upsampling can improve robot perception by ameliorating the 3D data quality.

关键词： deep learning for visual perception computer vision for automation

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-Assisted Image-Processing for Human-Robot Close Collaboration

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第2期7卷 4236-4243页

作者： Sajedi, Seyedomid Liu, Wansong Eltouny, Kareem Behdad, Sara Zheng, Minghui Liang, Xiao Univ Buffalo Civil Struct & Environm Engn Dept Buffalo NY 14260 USA Univ Buffalo Mech & Aerosp Engn Dept Buffalo NY 14260 USA Univ Florida Environm Engn Sci Dept Gainesville FL 32611 USA

The safety of human workers has been the main concern in human-robot close collaboration. Along with rapidly developed artificial intelligence techniques, deep learning models using two-dimensional images have become feasible solutions for human motion detection. These models serve as "sensors" in the closed-loop system that involve humans and robots. Most existing methods that detect human motion using images do not consider the uncertainty from the deep learning model itself. The mappings established by deep learning models should not be taken blindly, and thus uncertainty should be a natural part of this type of sensor. In particular, model uncertainty should be explicitly quantified and incorporated into robot motion control to guarantee safety. With this motivation, to rigorously quantify the uncertainty of these "sensors", this letter proposes a probabilistic interpretation method and automatically provides a framework to benefit from a deep model's uncertainty. Experimental data from human-robot collaboration has been collected and used to validate the proposed method. A training strategy is proposed to efficiently train surrogate models that learn to refine the prediction of the main Bayesian models. The proposed framework is also compared with Ego hands benchmark showing a 4.7% increase in mIoU.

关键词： Uncertainty Bayes methods Training Neural networks Safety deep learning Collaboration Human-robot collaboration computer vision for automation deep learning for visual perception

来源：评论

学校读者我要写书评

暂无评论

Partial-to-Partial Point Generation Network for Point Cloud Completion

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第4期7卷 11990-11997页

作者： Zhang, Ziyu Yu, Yi Da, Feipeng Southeast Univ Sch Automat Nanjing 211189 Jiangsu Peoples R China

Point cloud completion aims at predicting dense complete 3D shapes from sparse incomplete point clouds captured from 3D sensors or scanners. It plays an essential role in various applications such as autonomous driving, 3D reconstruction, augmented reality, and robot navigation. Existing point cloud completion methods follow the encoder-decoder paradigm, in which the complete point clouds are recovered in a coarse-to-fine strategy. However, only using the global feature is difficult and will lead to blurring of the global structure and distortion of local details. To address this problem, we propose a novel Partial-to-Partial Point Generation Network (P(2)GNet), a learning-based approach for point cloud completion. In P(2)GNet, we use a feature disentangle encoder to obtain the global feature, and missing code and novel view partial point cloud are generated conditioned on the view-related missing code. To better aggregate partial point clouds, an attentive sampling module is proposed to sample multiple partial point clouds into the final complete result. Extensive experiments on several public benchmarks demonstrate that our P(2)GNet outperforms state-of-the-art point cloud completion methods.

关键词： Computer Vision for Automation deep learning for visual perception Point Cloud Completion

来源：评论

学校读者我要写书评

暂无评论

visual Attention-Based Self-Supervised Absolute Depth Estimation Using Geometric Priors in Autonomous Driving

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第4期7卷 11998-12005页

作者： Xiang, Jie Wang, Yun An, Lifeng Liu, Haiyang Wang, Zijun Liu, Jian Chinese Acad Sci Inst Microelect Beijing 100029 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China

Although existing monocular depth estimation methods have made great progress, predicting an accurate absolute depth map from a single image is still challenging due to the limited modeling capacity of networks and the scale ambiguity issue. In this paper, we introduce a fully visual Attention-based Depth (VADepth) network, where spatial attention and channel attention are applied to all stages. By continuously extracting the dependencies of features along the spatial and channel dimensions over a long distance, VADepth network can effectively preserve important details and suppress interfering features to better perceive the scene structure for more accurate depth estimates. In addition, we utilize geometric priors to form scale constraints for scale-aware model training. Specifically, we construct a novel scale-aware loss using the distance between the camera and a plane fitted by the ground points corresponding to the pixels of the rectangular area in the bottom middle of the image. Experimental results on the KITTI dataset show that this architecture achieves the state-of-the-art performance and our method can directly output absolute depth without post-processing. Moreover, our experiments on the SeasonDepth dataset also demonstrate the robustness of our model to multiple unseen environments.

关键词： deep learning for visual perception computer vision for transportation range sensing

来源：评论

学校读者我要写书评

暂无评论

Cross-View and Cross-Domain Underwater Localization Based on Optical Aerial and Acoustic Underwater Images

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第2期7卷 4969-4974页

作者： Dos Santos, Matheus M. De Giacomo, Giovanni G. Drews-Jr, Paulo L. J. Botelho, Silvia S. C. Univ Fed Rio Grande FURG Ctr Computat Sci C3 Intelligent Robot & Automat Grp NAUTEC BR-96203900 Rio Grande Brazil

Cross-view image matches have been widely explored on terrestrial image localization using aerial images from drones or satellites. This study expands the cross-view image match idea and proposes a cross-domain and cross-view localization framework. The method identifies the correlation between color aerial images and underwater acoustic images to improve the localization of underwater vehicles that travel in partially structured environments such as harbors and marinas. The approach is validated on a real dataset acquired by an underwater vehicle in a marina. The results show an improvement in the localization when compared to the dead reckoning of the vehicle.

关键词： deep learning for visual perception marine robotics localization

来源：评论

学校读者我要写书评

暂无评论

Detaching and Boosting: Dual Engine for Scale-Invariant Self-Supervised Monocular Depth Estimation

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第4期7卷 12094-12101页

作者： Jiang, Peizhe Yang, Wei Ye, Xiaoqing Tan, Xiao Wu, Meng Northwestern Polytech Univ Sch Marine Sci & Technol Xian 710072 Peoples R China Baidu Inc Dept Comp Vis Technol VIS Beijing 100085 Peoples R China

Monocular depth estimation (MDE) in the self-supervised scenario has emerged as a promising method as it refrains from the requirement of ground truth depth. Despite continuous efforts, MDE is still sensitive to scale changes especially when all the training samples are from one single camera. Meanwhile, it deteriorates further since camera movement results in heavy coupling between the predicted depth and the scale change. In this paper, we present a scale-invariant approach for self-supervised MDE, in which scale-sensitive features (SSFs) are detached away while scale-invariant features (SIFs) are boosted further. To be specific, a simple but effective data augmentation by imitating camera zooming process is proposed to detach SSFs, making the model robust to scale changes. Besides, a dynamic cross-attention module is designed to boost SIFs by fusing multi-scale cross-attention features adaptively. Extensive experiments on the KITTI dataset demonstrate that the detaching and boosting strategies are mutually complementary in MDE and our approach achieves new State-of-The-Art performance against existing works from 0.097 to 0.090 w.r.t absolute relative error. The code will be made public soon.

关键词： Autonomous vehicle navigation deep learning for visual perception monocular depth estimation

来源：评论

学校读者我要写书评

暂无评论

On the Coupling of Depth and Egomotion Networks for Self-Supervised Structure from Motion

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2022年第3期7卷 6766-6773页

作者： Wagstaff, Brandon Peretroukhin, Valentin Kelly, Jonathan Univ Toronto Inst Aerosp Studies UTIAS Space & Terr Autonomous Robot Syst STARS Lab Toronto ON M3H 5T6 Canada MIT Comp Sci & Artificial Intelligence Lab Cambridge MA 02139 USA

Structure from motion (SfM) has recently been formulated as a self-supervised learning problem, where neural network models of depth and egomotion are learned jointly through view synthesis. Herein, we address the open problem of how to best couple, or link, the depth and egomotion network components, so that information such as a common scale factor can he shared between the networks. Towards this end, we introduce several notions of coupling, categorize existing approaches, and present a novel tightly-coupled approach that leverages the interdependence of depth and egomotion at training time and at test time. Our approach uses iterative view synthesis to recursively update the egomotion network input, permitting contextual information to be passed between the components. We demonstrate through substantial experiments that our approach promotes consistency between the depth and egomotion predictions at test time, improves generalization, and leads to state-of-the-art accuracy on indoor and outdoor depth and egomotion evaluation benchmarks.

关键词： deep learning for visual perception localization vision-based navigation

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：