中国科学院机构知识库网格系统: 检索

SANet: Statistic Attention Network for Video-Based Person Re-Identification 期刊论文 OAI收割
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2022, 卷号: 32, 期号: 6, 页码: 3866-3879
作者: Bai, Shutao; Ma, Bingpeng; Chang, Hong; Huang, Rui; Shan, Shiguang

|

收藏 | 浏览/下载：42/0 | 提交时间：2022/12/07

Feature extraction Task analysis Computational modeling Visualization Video sequences Fuses Computer science Person re-identification self-attention long-range dependencies high-order statistics

VidSfM: Robust and Accurate Structure-From-Motion for Monocular Videos 期刊论文 OAI收割
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2022, 卷号: 31, 页码: 2449-2462
作者: Cui, Hainan; Tu, Diantao; Tang, Fulin

|

收藏 | 浏览/下载：29/0 | 提交时间：2022/06/06

Cameras Image reconstruction Videos Simultaneous localization and mapping Video sequences Robustness Scalability Structure from motion image reconstruction computational geometry computer vision

Tangent Fisher Vector on Matrix Manifolds for Action Recognition 期刊论文 OAI收割
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2020, 卷号: 29, 页码: 3052-3064
作者: Luo, Guan; Wei, Jiutong; Hu, Weiming; Maybank, Stephen J.

|

收藏 | 浏览/下载：55/0 | 提交时间：2020/04/07

Manifolds Video sequences Observability Videos Covariance matrices Kernel Computational modeling Action recognition Fisher vector Grassmann manifold Hankel matrix matrix manifold

Applying maximally stable extremal regions and local binary patterns for guide-wire detecting in percutaneous coronary intervention 期刊论文 OAI收割
IET IMAGE PROCESSING, 2019, 卷号: 13, 期号: 13, 页码: 2579-2586
作者: Pusit, Prasong; Xie, Xiao-Liang; Hou, Zeng-Guang

|

收藏 | 浏览/下载：52/0 | 提交时间：2020/03/30

blood vessels medical image processing surgery image sequences video signal processing image filtering object detection X-ray imaging object tracking stroke width variation filter region detection local binary patterns guide-wire recognition conventional MSER methods maximally stable extremal regions guide-wire position anatomical skeleton contours training data X-ray video sequence percutaneous coronary intervention surgery region area range filter X-ray video monitoring guide-wire tip detection modified multifilters training templates

Validation of MODIS aerosol optical depth retrieval over mountains in central China based on a sun-sky radiometer site of SONET 期刊论文 OAI收割
Remote Sensing, 2016, 卷号: 8, 期号: 2
作者: Ma, Yan; Li, Zhengqiang; Li, Zhaozhou; Xie, Yisong; Fu, Qiaoyan

收藏 | 浏览/下载：39/0 | 提交时间：2017/04/24

OBJECT DETECTION VIDEO SEQUENCES KERNEL FEATURES

A simple and fast moving object segmentation based on H.264 compressed domain information (EI CONFERENCE) 会议论文 OAI收割
4th International Conference on Computational and Information Sciences, ICCIS 2012, August 17, 2012 - August 19, 2012, Chongqing, China
作者: Chen X.; Chen X.; Chen X.; Sun L.

收藏 | 浏览/下载：15/0 | 提交时间：2013/03/25

The paper presents a simple and fast approach for moving object segmentation based on H.264 compressed domain information for the application of indoor video surveillance with static camera. Due to the characteristics of indoor video surveillance the proposed method of segmentation avoids complicated background model like Gaussian Mixture background model. On the contrary it chooses some simple information like the type of Macroblock etc.. Experimental results of several specific H.264 compressed video sequences demonstrate the good segmentation quality of the proposed approach. 2012 IEEE.

Efficient human action recognition using accumulated motion image and support vector machines (EI CONFERENCE) 会议论文 OAI收割
International Workshop on Advanced Computational Intelligence and Intelligent Informatics, IWACIII 2011, November 19, 2011 - November 23, 2011, Suzhou, China
作者: Zhang X.; Zhang J.; Zhang J.; Zhang X.; Zhang X.

收藏 | 浏览/下载：68/0 | 提交时间：2013/03/25

Vision-based human action recognition provides an advanced interface and research in this field of human action recognition has been actively carried out. This paper describes a scheme for recognizing human actions from a video sequences. The proposed method is an extension of the Motion History Image(MHI) method based on the ordinal measure of accumulated motion which is robust to variations of appearances. We define the accumulated motion image(AMI) using image differences firstly. Then the AMI of the video sequencesis resized to a MN regulation following the standard of training phases. Finally we employ Support Vector Machine(SVM) as a classifier to distinguish the current activity in target video sequences. In a word our proposed algorithm not only outperforms the state of art on public available KTH data set and Weizmann data set but also proves practical to some real world applications in addition this method is computationally simple and able to achieve a satisfactory accuracy.

Optimization on motion estimation algorithm based on H. 264 (EI CONFERENCE) 会议论文 OAI收割
2010 3rd International Conference on Advanced Computer Theory and Engineering, ICACTE 2010, August 20, 2010 - August 22, 2010, Chengdu, China
Wen X.; Li G.

收藏 | 浏览/下载：56/0 | 提交时间：2013/03/25

Motion estimation is a very important part of video compression. As a result of using precision of motion vector in H.264 encoder the computational cost increases rapidly and motion estimation is the most time-consuming stage. In this paper based on the UMHexagonS algorithm an optimized algorithm is proposed based on the dynamic search window selection big hexagon and small hexagon search mode respectively which saves motion estimation time effectively with a little quality loss. Experiments with some typical video sequences show that compared to the original UMHexagonS algorithm this new algorithm can save about 17.851 % motion estimation time and reduce the complexity of original scheme as well as enhance the real time performance of encoder and almost has no changes in the reconstructed picture quality and bitrates. 2010 IEEE.

Adaptive deformation estimation of moving target by weight image analysis (EI CONFERENCE) 会议论文 OAI收割
2010 2nd International Conference on Future Computer and Communication, ICFCC 2010, May 21, 2010 - May 24, 2010, Wuhan, China
Bai X.-G.; Dai M.

收藏 | 浏览/下载：27/0 | 提交时间：2013/03/25

An algorithm based on weight image analysis is proposed for adaptive deformation estimation of moving target in mean-shift tracking method. At the first we get the weight image from the target candidate region. Then we analyze the differences between the object and background. According to that the area estimation of the target can be converted into the image segmentation task. To realize the adaptive segmentation and estimation we define the threshold as the maximum variance between object and background. Combining the estimated area and covariance matrix we can estimate the width height and orientation of the object. The experimental results on three representative video sequences validate its robustness to the deformable estimation of the targets. 2010 IEEE.

Super-resolution using adaptive blur parameter estimation (EI CONFERENCE) 会议论文 OAI收割
2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010, September 23, 2010 - September 25, 2010, Chengdu, China
作者: Wang H.; Wang H.; Wang H.; Wang H.; Liu G.

收藏 | 浏览/下载：16/0 | 提交时间：2013/03/25

Super-resolution is a term for a set of methods of increasing image or video resolution. All these methods are based on the same idea: using information from several images to create one upsized image. In most of the super-resolution algorithms the blur parameter of a LR-image model is always manually set as a default value. In this paper we propose a method to adaptively estimate the blur parameter. We get the initial image of iteration by fusing all low-resolution images. When it is used in MAP algorithm three iterations are enough to get a stable solution. It is greatly reduce the computational power compared with other MAP algorithms. Experiments to real image sequences show that it well preserved the image detail and the reconstructed image is clear. 2010 IEEE.