检索结果-内蒙古大学图书馆

2nd asia conference on computer vision, image processing and pattern recognition (cvippr)

作者： Li, WeiJia Chen, LiYuan Cai, JunZhi Tong, Lihan Chen, ErKang Huang, Bin Jimei Univ Xiamen Fujian Peoples R China

ISBN: (纸本)9798400716607

In this paper, we introduced NanoTrack, a novel multi-object tracking (MOT) method that leverages light-weight object detectors to enhance tracking performance in real-time applications where computational resources are scarce. While light-weight detectors are efficient, they often produce an imbalance in detection quality, generating a significant number of low-scoring detections that pose challenges for tracking algorithms. Our approach innovatively utilizes these low-scoring detections for track initialization and maintenance, addressing the shortcomings observed in existing tracking by two-stage tracking methods like ByteTrack, which struggle with the abundance of low-scoring detections. By integrating two new light-weight modules, Refind High Detection (RHD) and Duplicate Track Checking (DTC), NanoTrack effectively incorporates low-scoring detections into the tracking process. Additionally, we enhance the pseudo-depth estimation technique for improved handling in dense target environments, mitigating issues like ID Switching. Our comprehensive experiments demonstrate that NanoTrack surpasses state-of-the-art two-stage TBD methods, including ByteTrack and SparseTrack, on benchmark datasets such as MOT16, MOT17, and MOT20, thereby establishing a new standard for MOT performance using light-weight detectors. The code is open source in https://***/VjiaLi/NanoTrack

关键词： multi-object tracking computer vision light-weight object detector

来源：评论

学校读者我要写书评

暂无评论

A vision-based Remote Assistance Method and it's Application in Object Transfer 2

A Vision-based Remote Assistance Method and it's Application...

引用

2nd asia conference on computer vision, image processing and pattern recognition (cvippr)

作者： Cheng, Mingkai Yi, Pengfei Guo, Yujie Liu, Rui Dong, Jing Zhou, Dongsheng Dalian Univ Sch Software Engn Key Lab Adv Design & Intelligent Comp Minist Educ Dalian Liao Ning Peoples R China Dalian Univ Technol Sch Comp Sci & Technol Dalian Liao Ning Peoples R China

ISBN: (纸本)9798400716607

Remote assistance by users is a common means in current robot applications, which can improve the efficiency of transfer tasks. However, the current mainstream methods are costly, relying on auxiliary equipment, and result in a significant workload for users throughout the task. To reduce the cost and workload of remote assistance and improve user interaction with robots. This paper has undertaken the following work using available visual image information. Firstly, proposing a remote guidance method based on facial information, which is derived from analyzing user facial information for gaze estimation and combined with image detection. Secondly, proposing an autonomous transfer method based on position inference, which combines target information in images with the real-time state of the robot. Finally, by integrating the above methods, proposing a vision-based remote assistance method and applyingit in real-world experiments of object transfer tasks.

关键词： Visual processing Gaze Estimation Position Inference Object Transfer Remote Assistance

来源：评论

学校读者我要写书评

暂无评论

EDO-SANet: Shape-Aware Network with Edge Detection Operator for Polyp Segmentation 24

EDO-SANet: Shape-Aware Network with Edge Detection Operator ...

引用

2nd asia conference on computer vision, image processing and pattern recognition (cvippr)

作者： Zhao, Yifei Wang, Xiaoying Li, Yang Yin, Junping North China Elect Power Univ Beijing Peoples R China Northeast Normal Univ Changchun Peoples R China Inst Appl Phys & Computat Math Beijing Peoples R China

ISBN: (纸本)9798400716607

Automatic and accurate segmentation of colonic polyps can effectively help physicians quickly identify the size and location of polyps during endoscopy. However, due to the variable intestinal environment, polyps are highly similar to the surrounding tissue leading to over-segmentation or under-segmentation. To tackle these concerns, we propose a novel network with shape-awareness. Specifically, the network employs a dual-stream encoder, including the mainstream Transoformer-based encoder and an auxiliary encoder to extract global and local representations. The auxiliary encoder is built on deformable convolutional v3 for sensing the shape of polyps. In order to better identify the edges of polyps, the Edge Detection Guided Module (EDGM) is proposed to utilize edge detection operators to highlight the importance of edge information in low-level features. Furthermore, we introduce a Semantic Interaction Module (SIM) to better localize and calibrate targets by integrating global and local semantics in high-level features. Compared to other state-of-the-art methods, our model achieves better performance both in learning and generalization ability on five polyp segmentation datasets. The potential applicability of our model to other biomedical fields is demonstrated through its outstanding performance. The code is available at https://***/xff12138/EDOSANet

关键词： Polyp segmentation Medical image processing Deep learning Deformable convolution Edge detection operator

来源：评论

学校读者我要写书评

暂无评论

Research on Model-Free 6D Object Pose Estimation Based on vision 3D Matching 2

Research on Model-Free 6D Object Pose Estimation Based on Vi...

引用

2nd asia conference on computer vision, image processing and pattern recognition (cvippr)

作者： Chen, Yan Yi, Pengfei Guo, Yujie Liu, Rui Dong, Jing Zhou, Dongsheng Dalian Univ Sch Software Engn Key Lab Adv Design & Intelligent Comp Minist Educ Dalian Liao Ning Peoples R China Dalian Univ Technol Sch Comp Sci & Technol Dalian Liao Ning Peoples R China

ISBN: (纸本)9798400716607

6D object pose estimation is a crucial and fundamental task in the field of human-robot interaction. Existing visual target localization methods either need to rely on an additional depth camera to obtain spatial information of the target object or need to resort to a specific CAD model, which leads to high cost and poor adaptability of pose estimation. In this paper, we propose a 6D pose estimation method for objects based on RGB images and without CAD models, starting from the 3D reconstruction of objects. First, this paper constructs a sparse model of the target object by simple RGB scanning, then, 2D key points in the image are matched with 3D points in the object model using a feature-matching network, and finally, to alleviate the poor matching of such methods on low-resolution images, we introduce a 2D-3D-3D key points matching method, which achieves efficient and robust object 6D pose estimation results. Experimental results on the Onepose dataset demonstrate the accuracy and robustness of the method.

关键词： 6D Object Pose Estimation Key Points Matching RGB Modeling

来源：评论

学校读者我要写书评

暂无评论

Unifying Correspondence, Pose and NeRF for Generalized Pose-Free Novel View Synthesis

Unifying Correspondence, Pose and NeRF for Generalized Pose-...

引用

IEEE/CVF conference on computer vision and pattern recognition (CVPR)

作者： Hong, Sunghwan Jung, Jaewoo Shin, Heeseong Yang, Jiaolong Kim, Seungryong Luo, Chong Korea Univ Seoul South Korea Microsoft Res Asia Beijing Peoples R China

ISBN: (纸本)9798350353006

This work delves into the task of pose-free novel view synthesis from stereo pairs, a challenging and pioneering task in 3D vision. Our innovative framework, unlike any before, seamlessly integrates 2D correspondence matching, camera pose estimation, and NeRF rendering, fostering a synergistic enhancement of these tasks. We achieve this through designing an architecture that utilizes a shared representation, which serves as a foundation for enhanced 3D geometry understanding. Capitalizing on the inherent interplay between the tasks, our unified framework is trained end-to-end with the proposed training strategy to improve overall model accuracy. Through extensive evaluations across diverse indoor and outdoor scenes from two real-world datasets, we demonstrate that our approach achieves substantial improvement over previous methodologies, especially in scenarios characterized by extreme viewpoint changes and the absence of accurate camera poses. The project page and code will be made available at: https://***/CoPoNeRF/.

关键词： Stereo image processing

来源：评论

学校读者我要写书评

暂无评论

image recognition and Analysis: Results of Belarusian Scientists

引用

pattern recognition AND image ANALYSIS 2023年第4期33卷 626-645页

作者： Ablameyko, S. V. Belarusian State Univ Minsk 220030 BELARUS Problems Natl Acad Sci Belarus United Inst Informat Minsk 220012 BELARUS

The article describes the results of Belarusian scientists in the field of image recognition and analysis obtained over the last 30 years. The main groups of scientists working at the National Academy of Sciences of Belarus and universities are shown. conferences held by Belarusian scientists are shown, with special attention paid to the most popular conference pattern recognition and Information processing. Information is provided on the development of the knowledge-intensive IT sector and companies working in the field of computer vision. Data on the training of scientific personnel in this field are provided.

关键词： image recognition and analysis computer vision results of Belarusian scientists

来源：评论

学校读者我要写书评

暂无评论

A Method of Yoga Action pattern recognition Based on computer vision Technology

A Method of Yoga Action Pattern Recognition Based on Compute...

引用

2023 asia-Europe conference on Electronics, Data processing and Informatics, ACEDPI 2023

作者： Ma, Ling Huang, Chao Wenhua College Hubei Wuhan430074 China

ISBN: (纸本)9798350301151

Object detection and recognition using computer vision has been a very interesting and challenging research field in the past three decades. Classification and target location based on machine learning and computer vision technology has always been a hot topic, and great achievements have been made. Firstly, the classification analysis method is used to analyze and study the yoga movement and theory based on computer vision technology, including human eye tracking and recognition, face recognition, head movement tracking and recognition, gesture recognition and posture recognition. Based on the task of yoga movement pattern recognition, a yoga movement pattern recognition method based on computer vision technology is proposed. The improved model is based on the framework and structure of the network. A certain number of candidate regions are proposed and classified through feature extraction, and then these regions are output as the detected bounding box. The posture action diagram of yoga, that is, ordinary RGB color image, is collected by RGB camera, and the bone data is extracted from RGB image by bone extraction model. The RGB image data and bone data are input into the joint model, and the joint model will output the category of Yoga action and the score of this action. © 2023 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Medical image Segmentation Approach via Transformer Knowledge Distillation 23

Medical Image Segmentation Approach via Transformer Knowledg...

引用

2023 asia conference on computer vision, image processing and pattern recognition, cvippr 2023

作者： Zhang, Tian-Tian Wang, Hao Lam, Kam-Yiu Chow, Chi-Yin Department of Computer Science City University of Hong Kong Hong Kong Department of Electrical Engineering City University of Hong Kong Hong Kong Limited Hong Kong

ISBN: (纸本)9798400700033

Numerous transformer-based medical image segmentation methods have been proposed and achieved good segmentation results. However, it is still a challenge to train and deploy transformer networks to mobile medical devices due to a large number of model parameters. To resolve the training and model parameter problems, in this paper, we propose a Transformer-based network for Medical image Segmentation using Knowledge Distillation named MISTKD. The MISTKD consists of a teacher network and a student network. It achieves comparable performance to state-of-the-art transformer works using fewer parameters by employing the teacher network to train the student network. The training can be implemented by extracting the sequence in the teacher and student encoder networks during the training procedure. The losses between sequences are further calculated, thus the student network can learn from the teacher network. The experimental results on Synapse show that the proposed work achieves competitive performance using only one-eighth parameters. © 2023 ACM.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

3D Ultrasound Tomography image Reconstruction Algorithm by GPU 23

3D Ultrasound Tomography Image Reconstruction Algorithm by G...

引用

2023 asia conference on computer vision, image processing and pattern recognition, cvippr 2023

作者： Gong, Jiaduo Donghua University China

ISBN: (纸本)9798400700033

At present, X-ray technology, B-ultrasound and magnetic resonance imaging technology have more or less defects in the detection of female breast cancer, so the early detection of breast cancer is still a very important challenge. Ultrasound tomography (UT) can solve these problems very well. This project mainly uses the TVAL3 algorithm to reconstruct the original image from the information collected by the UT system for clinical use. TVAL3 algorithm involves a large number of matrix-vector multiplications and transposed matrix-vector multiplications, which will consume a lot of time if traditional CPU methods are used. For the characteristics of matrix-vector multiplication, this project uses CUDA to call GPU for parallel computing. At the same time, in order to further increase the speed of the calculation, we put part of the unchanged content into the GPU in advance to reduce the time spent on the transfer process. The final speedups of 20x, 10x and 5x were achieved in matrix vector multiplication, transpose matrix vector multiplication and total time, respectively. © 2023 ACM.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Human Pose as Compositional Tokens

Human Pose as Compositional Tokens

引用

IEEE/CVF conference on computer vision and pattern recognition (CVPR)

作者： Geng, Zigang Wang, Chunyu Wei, Yixuan Liu, Ze Li, Houqiang Hu, Han Univ Sci & Technol China Hefei Peoples R China Tsinghua Univ Beijing Peoples R China Microsoft Res Asia Beijing Peoples R China

ISBN: (纸本)9798350301298

Human pose is typically represented by a coordinate vector of body joints or their heatmap embeddings. While easy for data processing, unrealistic pose estimates are admitted due to the lack of dependency modeling between the body joints. In this paper, we present a structured representation, named Pose as Compositional Tokens (PCT), to explore the joint dependency. It represents a pose by M discrete tokens with each characterizing a sub-structure with several interdependent joints (see Figure 1). The compositional design enables it to achieve a small reconstruction error at a low cost. Then we cast pose estimation as a classification task. In particular, we learn a classifier to predict the categories of the M tokens from an image. A pre-learned decoder network is used to recover the pose from the tokens without further post-processing. We show that it achieves better or comparable pose estimation results as the existing methods in general scenarios, yet continues to work well when occlusion occurs, which is ubiquitous in practice. The code and models are publicly available at https://***/Gengzigang/PCT.

关键词： body gesture Humans: Face movement pose

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：