检索结果-内蒙古大学图书馆

IEEE Applied Imagery Pattern Recognition Workshop (AIPR)

作者： Lee, Yaesop Lee, Hyungtae Lee, Eungjoo Kwon, Heesung Bhattacharyya, Shuvra Univ Maryland Dept ECE College Pk MD 20742 USA Univ Maryland Inst Adv Comp Studies College Pk MD 20742 USA US Army Res Lab Intelligent Percept Branch Adelphi MD USA MGH Dept Radiol CAMCA Boston MA USA Harvard Med Sch Boston MA USA

ISBN: (纸本)9781665477291

Stereo image inputs provide higher object detection accuracy than monocular images by enabling the detection of objects that are missed from one view while being detectable from another view. To take advantage of additional information from the secondary image, it is necessary to search for the corresponding region in the images of different views by projecting with depth information of the target object. However, most existing studies utilize highly complex computations to estimate the depth for simple 2d object detection. This complexity limits the potential for deploying the methods on platforms, such as unmanned aerial vehicles, that involve significant resource constraints. In this paper, we introduce a simplified depth approximation to obtain depth information by quantizing the depth values into a small number of representative values. With these values, the regions of interest are projected to the secondary image to concatenate the information from the additional image. We validate our method with the KITTI dataset. Our results show that while having very low complexity, our approximation method leads to greatly improved object detection performance in two out of three difficulty groups of the dataset, and comparable performance in the other difficulty group compared to use of monocular image input.

关键词： 2d object detection stereo depth estimation

来源：评论

学校读者我要写书评

暂无评论

Joint 2d object detection and 3d Reconstruction via Adversarial Fusion Mesh R-CNN 53

Joint 2D Object Detection and 3D Reconstruction via Adversar...

引用

IEEE International Symposium on Circuits and Systems (IEEE ISCAS)

作者： Zhou, Zihan Lai, Qinghan ding, Shuai Liu, Song Qilu Univ Technol Sch Comp Sci & Technol Shandong Acad Sci Jinan Peoples R China

ISBN: (纸本)9781728192017

Joint 2d object detection and 3d reconstruction is an essential computer vision task to get more accurate detection and representation model of the target object. We proposed a novel joint 2d object detection and 3d reconstruction model that enhances the ability of the 2d object detection and the 3d reconstruction, called Adversarial Fusion Mesh Region Convolutional Neural Networks (AFM R-CNN). Our proposed model introduces the deep Convolutional Generative Adversarial Network (dCGAN) to generate adversarial images and input the real and adversarial images into the object detection module GA-RPN to determine the position and anchor box of the target object. Next, to make better use of the two-dimensional information of the image, the voxel conversion and Fusion model Pix2Vox is introduced to fuse the two types of image features and generate coarse voxels. Afterwards, to differentiate the voxel information more efficiently, we use the Principal Neighborhood Aggregation network (PNA) model in 3d model refinement. The contrast experimental results on the open domain dataset (Pix3d) with baseline models demonstrate the effectiveness of AFM R-CNN in joint 2d object detection and 3d reconstruction task.

关键词： 2d object detection 3d Reconstruction Voxel Conversion and Fusion 3d Model Refinement

来源：评论

学校读者我要写书评

暂无评论

A Survey of Computer Vision Methods for 2d object detection from Unmanned Aerial Vehicles

引用

JOURNAL OF IMAGING 2020年第8期6卷 78-78页

作者： Cazzato, dario Cimarelli, Claudio Sanchez-Lopez, Jose Luis Voos, Holger Leo, Marco Univ Luxembourg Interdisciplinary Ctr Secur Reliabil & Trust SnT L-1855 Luxembourg Luxembourg Natl Res Council Italy Inst Appl Sci & Intelligent Syst I-73100 Lecce Italy

The spread of Unmanned Aerial Vehicles (UAVs) in the last decade revolutionized many applications fields. Most investigated research topics focus on increasing autonomy during operational campaigns, environmental monitoring, surveillance, maps, and labeling. To achieve such complex goals, a high-level module is exploited to build semantic knowledge leveraging the outputs of the low-level module that takes data acquired from multiple sensors and extracts information concerning what is sensed. All in all, the detection of the objects is undoubtedly the most important low-level task, and the most employed sensors to accomplish it are by far RGB cameras due to costs, dimensions, and the wide literature on RGB-based object detection. This survey presents recent advancements in 2d object detection for the case of UAVs, focusing on the differences, strategies, and trade-offs between the generic problem of object detection, and the adaptation of such solutions for operations of the UAV. Moreover, a new taxonomy that considers different heights intervals and driven by the methodological approaches introduced by the works in the state of the art instead of hardware, physical and/or technological constraints is proposed.

关键词： computer vision 2d object detection unmanned aerial vehicles deep learning

来源：评论

学校读者我要写书评

暂无评论

CenterNet-Auto: A Multi-object Visual detection Algorithm for Autonomous driving Scenes Based on Improved CenterNet

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2023年第3期7卷 742-752页

作者： Wang, Hai Xu, Yansong Wang, Zining Cai, Yingfeng Chen, Long Li, Yicheng Jiangsu Univ Sch Automot & Traff Engn Zhenjiang 212013 Peoples R China Jiangsu Univ Automot Engn Res Inst Zhenjiang 212013 Peoples R China

With the rise in popularity of autonomous driving, the speed and accuracy of surrounding objects' detection by in-vehicle sensing technology is becoming increasingly important for autonomous vehicles. Building on CenterNet, this paper proposes CenterNet-Auto, a new anchor-free detection network for driving scenes that can satisfy the detection speed requirements while ensuring detection accuracy. The network's backbone uses the RepVGG model transformed through structural re-parameterization technology. Features of different scales are fused, and feature pyramids and deformable convolution are added after the backbone to accurately detect objects of different sizes. To solve the occlusion problem in the driving scene, this paper proposes the Average Border Model, which supports locating the object using the boundary feature information. The test results demonstrate that the proposed algorithm outperforms CenterNet regarding speed and accuracy on the Bdd dataset. The accuracy reaches 55.6%, and the speed reaches 30 FPS, meeting the speed and accuracy requirements in a driving scene.

关键词： Feature extraction detection algorithms Convolution object detection Head Proposals Training 2d object detection autonomous driving complex traffic conditions image recognition

来源：评论

学校读者我要写书评

暂无评论

Real-Time Monocular Joint Perception Network for Autonomous driving

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2022年第9期23卷 15864-15877页

作者： Li, Keqiang Xiong, Hui Liu, Jinxin Xu, Qing Wang, Jianqiang Tsinghua Univ State Key Lab Automot Safety & Energy Sch Vehicle & Mobil Beijing 100084 Peoples R China

Comprehensive and accurate perception of the real 3d world is the basis of autonomous driving. However, many perceptual methods focus on a single task or object type, and the accuracy of existing multi-task or multi-object methods is difficult to balance against their real-time performance. This paper presents a unified framework for concurrent dynamic multi-object joint perception, which introduces a real-time monocular joint perception network termed MJPNet. In MJPNet relative weightings are automatically learned by a series of developed network branches. By training an end-to-end deep convolutional neural network on a shared feature encoder and many proposed decoding sub-branches, the information of the 2d category and 3d position/pose/size of an object are reconstructed both simultaneously and accurately. Moreover, the effective information among subtasks is transferred by multi-stream learning, guaranteeing the accuracy of each task. Compared to various state-of-the-arts, comprehensive evaluations on the benchmark of challenging image sequences demonstrate the superior performance of our 2d detection and 3d reconstruction of depth, lateral distance, orientation, and heading angle. Moreover, on the KITTI test set, the real-time runtime (up to 15 fps) of MJPNet significantly outran the public state-of-the-art visual detection methods. Accompanying video: https://***/Z-goToOlI94.

关键词： Three-dimensional displays Estimation Feature extraction Task analysis Real-time systems object detection Image reconstruction 2d object detection 3d object reconstruction deep neural networks depth estimation orientation estimation

来源：评论

学校读者我要写书评

暂无评论

Adversarial Attacks against Traffic Sign detection for Autonomous driving 7

Adversarial Attacks against Traffic Sign Detection for Auton...

引用

7th CAA International Conference on Vehicular Control and Intelligence, CVCI 2023

作者： Xu, Feiyang Li, Ying Yang, Chao Wang, Weida Xu, Bin School of Mechanical Engineering Beijing Institute of Technology Beijing China

ISBN: (纸本)9798350340488

deep neural networks play a crucial role in 2d object detection based on visual data, but they are also vulnerable to adversarial samples. Attackers manipulate low-resolution images to execute data poisoning attacks. This paper introduces a method to generate realistic high-resolution adversarial samples aimed at compromising traffic sign detection models. Specifically, we propose a high-resolution adversarial sample framework built upon generative adversarial networks. Subsequently, an adversarial traffic sign detection model is developed to investigate the impact of data poisoning. To enhance the model's robustness, we conduct adversarial training. Experimental results demonstrate the efficacy of our data poisoning approach in misleading the detection model. Furthermore, the detection model exhibits improved robustness against such attacks following adversarial training. © 2023 IEEE.

关键词： 2d object detection data poisoning generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Head design for object detectors 25

Hierarchical Head Design for Object Detectors

引用

25th International Conference on Pattern Recognition (ICPR)

作者： Agarwal, Shivang Jurie, Frederic Normandie Univ CNRS ENSICAEN UNICAEN Caen France

ISBN: (纸本)9781728188089

The notion of anchor plays a major role in modern detection algorithms such as the Faster-RCNN or the SSd detector [2]. Anchors relate the features of the last layers of the detector with bounding boxes containing objects in images. despite their importance, the literature on object detection has not paid real attention to them. The motivation of this paper comes from the observations that (i) each anchor learns to classify and regress candidate objects independently (ii) insufficient examples are available for each anchor in case of small-scale datasets. This paper addresses these questions by proposing a novel hierarchical head for the SSd detector. The new design has the added advantage of no extra weights, as compared to the original design at inference time, while improving detectors performance for small size training sets. Improved performance on PASCAL-VOC and state-of-the-art performance on FlickrLogos-47 validate the method. We also show when the proposed design does not give additional performance gain over the original design.

关键词： 2d object detection Computer Vision Anchors deep Learning

来源：评论

学校读者我要写书评

暂无评论

Multiscale object detection from drone Imagery Using Ensemble Transfer Learning

引用

dRONES 2021年第3期5卷 66-66页

作者： Walambe, Rahee Marathe, Aboli Kotecha, Ketan Symbiosis Int Deemed Univ SIU Symbiosis Ctr Appl Artificial Intelligence SCAAI Pune 412115 Maharashtra India Symbiosis Int Deemed Univ SIU Symbiosis Inst Technol Pune 412115 Maharashtra India Savitribai Phule Pune Univ Pune Inst Comp Technol Pune 411043 Maharashtra India

object detection in uncrewed aerial vehicle (UAV) images has been a longstanding challenge in the field of computer vision. Specifically, object detection in drone images is a complex task due to objects of various scales such as humans, buildings, water bodies, and hills. In this paper, we present an implementation of ensemble transfer learning to enhance the performance of the base models for multiscale object detection in drone imagery. Combined with a test-time augmentation pipeline, the algorithm combines different models and applies voting strategies to detect objects of various scales in UAV images. The data augmentation also presents a solution to the deficiency of drone image datasets. We experimented with two specific datasets in the open domain: the Visdrone dataset and the AU-AIR dataset. Our approach is more practical and efficient due to the use of transfer learning and two-level voting strategy ensemble instead of training custom models on entire datasets. The experimentation shows significant improvement in the mAP for both Visdrone and AU-AIR datasets by employing the ensemble transfer learning method. Furthermore, the utilization of voting strategies further increases the 3reliability of the ensemble as the end-user can select and trace the effects of the mechanism for bounding box predictions.

关键词： drone imagery 2d object detection ensemble techniques voting strategies

来源：评论

学校读者我要写书评

暂无评论

An Optimized Multi-sensor Fused object detection Method for Intelligent Vehicles 5

An Optimized Multi-sensor Fused Object Detection Method for ...

引用

IEEE 5th International Conference on Intelligent Transportation Engineering (ICITE)

作者： Shen, Jiayu Liu, Qingxiao Chen, Huiyan Beijing Inst Technol Intelligent Vehicle Res Ctr Beijing Peoples R China

ISBN: (纸本)9781728194097

An accurate and efficient environment perception system is crucial for intelligent vehicles. This study proposes an optimized 2d object detection method utilizing multi-sensor fusion to improve the performance of the environment perception system. In the sensor fusion module, a depth completion network is used to predict dense depth map, so both dense and sparse RGB-d images can be obtained. Then, an efficient object detection baseline is optimized for intelligent vehicles. This method is verified by KITTI 2d object detection dataset. The experimental results show that the proposed method can be more accurate than many latest methods on KITTI leaderboard. Meanwhile, this method consumes less inference time and shows its high efficiency.

关键词： 2d object detection multi-sensor fusion deep learning

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：