Abstract:This paper proposes a practical multiple object tracking system based on DeepSort (Simple Online and Realtime Tracking with a Deep Association Metric). This system uses the tracking-by-detection method as the framework, which divides the optimization processes into two parts: detection and tracking. As for the overall performance of the algorithm, detection quality is a key element for it. According to the comparison in terms of speed and performance, suitable detectors are SDP-CRC and Yolov3 in multi-object tracking scenes. Additionally, this project has proposed a new pre-processing method based on NMS (non maximum suppression). This method has improved tracking performance by up to 3.5\% in terms of MOTA. As for tracking components, this project replaces a new estimation method based on visual object trackers (SiamRPN) with traditional Kalman motion estimation method. The experimental results show that this method has improved MOTA (Multiple Object Tracking Accuracy) by 0.3\%. Besides, the speed performance has decreased nearly 500\%. That’s because this experiment should be carried out more finely. In the next step, the performance of this method can be improved by updating the template frame appropriately, using appearance information provided by SiamRPN tracker for matching process and improving the robustness of the tracker. Additionally, this project has optimized detection-to-tracker association algorithm by using their positional and velocity relationship. It has reduced the complexity of the algorithm by avoiding multi-dimensional matrix operations.

Alert Name:
Alerting to:
Authentication email will be sent to your email address in 24 hours
Frequency:
Email Message Format:	Plain text Graphical(HTML)

Complete the form below and we will recommend the selected titles to your friends on your behalf. * Indicates a required field.
Your name*:
Your email address*:
Recipient's name*:
Recipient's email address*:
(multiple recipient's names and email addresses should be separated with semicolons)
Your comments:	I thought you would find the page(s) useful.

Your name:
Your email address:
Recipient's name:
Recipient's email address:
(multiple recipient's names and email addresses should be separated with semicolons)
Your comments:	I thought you would find this page useful.

Disclaimer: This message was sent to your friend using the "Send it to a friend" facility on the Sciencepaper Online’ WWW site, http://www.paper.edu.cn/en. The Sciencepaper Online is not responsible for the content of this email, and anything said in this email does not necessarily reflect the Sciencepaper Online's views.


	1. Research on Multi-Object Tracking Algorithm Based on Deep Feature
	SUN Qing-Hong, DONG Yuan
	Computer Science and Technology 23 January 2022
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:This paper proposes a practical multiple object tracking system based on DeepSort (Simple Online and Realtime Tracking with a Deep Association Metric). This system uses the tracking-by-detection method as the framework, which divides the optimization processes into two parts: detection and tracking. As for the overall performance of the algorithm, detection quality is a key element for it. According to the comparison in terms of speed and performance, suitable detectors are SDP-CRC and Yolov3 in multi-object tracking scenes. Additionally, this project has proposed a new pre-processing method based on NMS (non maximum suppression). This method has improved tracking performance by up to 3.5\% in terms of MOTA. As for tracking components, this project replaces a new estimation method based on visual object trackers (SiamRPN) with traditional Kalman motion estimation method. The experimental results show that this method has improved MOTA (Multiple Object Tracking Accuracy) by 0.3\%. Besides, the speed performance has decreased nearly 500\%. That’s because this experiment should be carried out more finely. In the next step, the performance of this method can be improved by updating the template frame appropriately, using appearance information provided by SiamRPN tracker for matching process and improving the robustness of the tracker. Additionally, this project has optimized detection-to-tracker association algorithm by using their positional and velocity relationship. It has reduced the complexity of the algorithm by avoiding multi-dimensional matrix operations.
	TO cite this article:SUN Qing-Hong, DONG Yuan. Research on Multi-Object Tracking Algorithm Based on Deep Feature[OL].[23 January 2022] http://en.paper.edu.cn/en_releasepaper/content/4756184


	2. Efficient Image Inpainting with Knowledge Distillation
	Cheng Chuxuan,Shen Qiwei,Wang Jing
	Computer Science and Technology 10 January 2020
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:In recent years, deep learning has made outstanding achievements in image classification, recognition, segmentation and generation. And some breakthroughs have been made in the research of image inpainting based on deep learning.The existing algorithms workwell, but they cannot inference in real time.In order to realize fast and efficient image inpainting,optimization is carried out from three aspectsbased on gated convolution. Using the pyramid sample to optimize the dilated gating convolution layers and proposing a coarse-to-fine pyramid sampling network(PUNet), compared with the gating convolution network, PUNet has less computation and more parameters to learn characteristics, as well as integrating different depth characteristics. Proposing holistic，pair-wise，pixel-wise loss function to enhance the local and global consistency. Introducing knowledge distillation into image inpainting and designs a multi-level self-distillation method. Experiments show that PUNet achieves the similar performance to gated convolutional network with 22% inference time.
	TO cite this article:Cheng Chuxuan,Shen Qiwei,Wang Jing. Efficient Image Inpainting with Knowledge Distillation[OL].[10 January 2020] http://en.paper.edu.cn/en_releasepaper/content/4750420

Saved Papers

Saved Papers


	3. AN EFFICIENT DEEP LEARNING HASHING NEURAL NETWORK FOR MOBILE VISUAL
	Heng Qi,Liang Liu
	Computer Science and Technology 26 January 2018
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Mobile visual search applications are emerging that enable users to sense their surroundings with smart phones. However, because of the particular challenges of mobile visual search, achieving a high recognition bitrate has becomes a consistent target of previous related works. In this paper, we propose a few-parameter, low-latency, and high-accuracy deep hashing approach for constructing binary hash codes for mobile visual search. First, we exploit the architecture of the MobileNet model, which significantly decreases the latency of deep feature extraction by reducing the number of model parameters while maintaining accuracy. Second, we add a hash-like layer into MobileNet to train the model on labeled mobile visual data. Evaluations show that the proposed system can exceed state-of-the-art accuracy performance in terms of the MAP. More importantly, the memory consumption is much less than that of other deep learning models. The proposed method requires only 13 MB of memory for the neural network and achieves a MAP of 97.80% on the mobile location recognition dataset used for testing.
	TO cite this article:Heng Qi,Liang Liu. AN EFFICIENT DEEP LEARNING HASHING NEURAL NETWORK FOR MOBILE VISUAL[OL].[26 January 2018] http://en.paper.edu.cn/en_releasepaper/content/4743206


	4. A Light Eyetracking System for Webpage Designing
	Chen Daiwu,Zhang Honggang,Guo Jie,Zhang Nannan
	Computer Science and Technology 16 September 2014
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Eye-tracking technology has always being considered as one of the most important methods in various fields concerning human's behaviour and cognition. This paper targets on designing and implementing a cheap real-time eye-tracking system mainly depend on a web-camera and software processing for rising need of webpage designing. We will introduce new tracking pattern, new eye features sets and SVM(Support Vector Machine)in this system. And experiments are taken based on proposed method with satisfied precision.
	TO cite this article:Chen Daiwu,Zhang Honggang,Guo Jie, et al. A Light Eyetracking System for Webpage Designing[OL].[16 September 2014] http://en.paper.edu.cn/en_releasepaper/content/4610154


	5. Design of a system for classifying the relationship based on video behavior analysis
	Wu Hao
	Computer Science and Technology 12 February 2012
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Moving object detection and video behavior analysis has become one of the hotspots in modern society. Analysis and hazard identification is also on the foundation of it.Through moving object detection and classification algorithm ,we can dig up more information from the video.We do further research on the foundation of prior research and adopt Pathfinding algorithm , Bayesian estimation to classify the relationship between two people.The system can be used in supermarket ,office and some important place .It also can be used for investigation work.At the same time,the system can help us collect information,can be widely used in market analysis ,can be used to get more commercial value.
	TO cite this article:Wu Hao. Design of a system for classifying the relationship based on video behavior analysis[OL].[12 February 2012] http://en.paper.edu.cn/en_releasepaper/content/4465989


	6. Head Pose Estimation based on the Mahanalobis Sparse Representation Classifier
	MA Bingpeng,PANG Xiumei,HU Weijun
	Computer Science and Technology 05 August 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:In this paper, a novel method is presented to improve the accuracy of head pose estimation. For the classify in head pose, the Mahanalobis Sparse Representation Classifier~(MSRC) method is proposed to improve the ability of classify. By using the Mahanalobis distance of samples belongs to the different classes, MSRC improve the classify ability from reducing the reconstructed error of the testing samples. The combination of MSRC can improve the accuracy of head pose estimation greatly. To show the effectiveness of MSRC, we compared them with other methods under two different databases. The results of the experiments show the proposed methods can improve the accuracy of head pose estimation.
	TO cite this article:MA Bingpeng,PANG Xiumei,HU Weijun. Head Pose Estimation based on the Mahanalobis Sparse Representation Classifier[OL].[ 5 August 2011] http://en.paper.edu.cn/en_releasepaper/content/4437887


	7. Research on Fault Diagnosis System of Mine Ventilator Based on Elman Neural Network
	Ren Zihui,Li Jiangang,Liu Yanxia
	Computer Science and Technology 28 June 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:This paper introduced the theory, learning algorithm and technical route of Elman neural. Though acquainting fault signals on-site and normalizing characteristic data, this method realized intelligent diagnosis of ventilator by constructing optimum structure and parameters based on Elman neural network. Compared with the traditional BP neural network, Elman network had a better comprehensive performance in diagnosis of ventilator. The result for the fault diagnosis of a ventilator showed that the Elman network improves the study speed, represses the network to sink local minimum, shortens the study time, and Elman neural is a effective method for the fault diagnosis of ventilator.
	TO cite this article:Ren Zihui,Li Jiangang,Liu Yanxia. Research on Fault Diagnosis System of Mine Ventilator Based on Elman Neural Network[OL].[28 June 2011] http://en.paper.edu.cn/en_releasepaper/content/4433667


	8. Face Extraction from Video Sequences by K-means Clustering and Fusion
	JI Luping
	Computer Science and Technology 25 May 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Face segmentation is an important processing step in a typically face-based person identification system. This paper presents a modified K-means clustering and fusion approach for face region extraction from colorful images of video sequences. It is based on Lab* color space analysis and consists of three processing steps: color number estimation, color region clustering, and face region post-processing. Moreover, experimental results on CIPR public video sequences are also exhibited to verify the feasibility, high efficiency and accuracy.
	TO cite this article:JI Luping. Face Extraction from Video Sequences by K-means Clustering and Fusion[OL].[25 May 2011] http://en.paper.edu.cn/en_releasepaper/content/4429836


	9. An online music recognizing and indexing system
	Guanxiong Wang,Liu Gang
	Computer Science and Technology 27 February 2009
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:In this paper, we present a human tone based online music system. This system not only espouses the tradition key word retrieval, but also espouses the humming and fragment of a music input to find out the according music. Through the procedures of recognition, coding and indexing, this system could locate target files properly and efficiently. Comparing with the key word retrieval method, the system is more convenient for users, specially the forget the name, libretto, composer and so on.
	TO cite this article:Guanxiong Wang,Liu Gang . An online music recognizing and indexing system[OL].[27 February 2009] http://en.paper.edu.cn/en_releasepaper/content/29737


	10. An Improved Aggregated One-Dependence Estimator:On Not So Rigid Cross Validation
	Zheng Qinghua
	Computer Science and Technology 06 January 2008
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:1The naive Bayesian classifier is a simple and effective approach to classifier learning. However, its conditionally independent assumption doesn’t often hold in the real world and this will lead to accuracy decrease in some applications. LBR, TAN and AODE are several representative algorithms that seek to relax this assumption and have exhibited noticeable prediction performance. However, the computational cost of LBR and TAN is considerable. AODE is creditably an effective classification learning algorithm without increasing computational cost improperly. However, to intently debase computational complexity, AODE avoids model selection and adopts all SPODEs , this may result in insufficiency to improve the prediction accuracy because of the included SPODEs which bring negative effect. Therefore, we propose an improved algorithm in this paper which will improve classification accuracy and classification speed by filtering out those SPODEs which bring negative effect base on original algorithms.
	TO cite this article:Zheng Qinghua . An Improved Aggregated One-Dependence Estimator:On Not So Rigid Cross Validation[OL].[ 6 January 2008] http://en.paper.edu.cn/en_releasepaper/content/17765

	Check out RSS, or use RSS reader to subscribe this item