Description: Robust Non-negative Dictionary Learning for Visual Tracking
The provided codes could be either embedded into the benchmark framework of paper Online Object Tracking: A Benchmark (CVPR2013) (You can find details here: http://visual-tracking.net/) or run on individual sequence.
To run the benchmark, just put the entire folder into the /trackers folder in the benchmark code base, and modify the configTrackers.m in util folder. DLT gets an AUC of 0.436, which ranks 5th among 26 in the benchmark by 19/03/2014. We don t tune parameters for single sequence in this case, all the parameters are stored in trackparam_DLT.m.
To run on individual video, you need to modify the dataPath and title in run_individual.m.
If you run MATLAB version after 2012, and have a CUDA compatible GPU installed, you may enjoy the fast computation speed by GPU, just set useGPU to true in trackparam_DLT.m and run_individual.m!
-Robust Non-negative Dictionary Learning for Visual Tracking
The provided codes could be either embedded into the benchmark framework of paper Online Object Tracking: A Benchmark (CVPR2013) (You can find details here: http://visual-tracking.net/) or run on individual sequence.
To run the benchmark, just put the entire folder into the /trackers folder in the benchmark code base, and modify the configTrackers.m in util folder. DLT gets an AUC of 0.436, which ranks 5th among 26 in the benchmark by 19/03/2014. We don t tune parameters for single sequence in this case, all the parameters are stored in trackparam_DLT.m.
To run on individual video, you need to modify the dataPath and title in run_individual.m.
If you run MATLAB version after 2012, and have a CUDA compatible GPU installed, you may enjoy the fast computation speed by GPU, just set useGPU to true in trackparam_DLT.m and run_individual.m!
Platform: |
Size: 22211584 |
Author:mohit |
Hits:
Description: 智能图像/视频处理中，复杂背景环境（比如室外环境、机场、车站等）下，人脸识别的第一步是人脸的检测。它的精确度直接影响到后期识别的结果。不过，领域内的科学家们基本上很难有足够的精力和时间开发优化的C++代码，使其用于商业用途，而一般都是只在Matlab中进行模拟。
本文的目的是提供一个我开发的SSE优化的，C++库，用于人脸检测，你可以马上把它用于你的视频监控系统中。文章中的分类器的训练数据来自与我的 webcam图像，它们被采集于不同时间，不同光照，不同背景环境下，它几乎可以实时地检测出我（的脸：）。训练的非人脸数据来自对不同背景的采集，用的是同一个webcam。被提取出的人脸区域，已经经过下面的处理：高斯滤波，直方图均衡化。
如果你需要更精确的结果，请从internet上下载更多不同的人脸集合，然后从新训练分类器。和我的库中一样尺寸的公共库是CBCL，其库超过100MB，所以，请大家自己下载楼-Intelligent image/video processing, complex background environment (such as an outdoor environment, airports, stations, etc.), the first step is the recognition of face detection. It directly affects the accuracy of the latter part of the identification results. However, scientists in the field are basically difficult to have enough energy and time to develop optimized C++ code to be used for commercial purposes, and are generally only be simulated in Matlab. The purpose of this paper is to provide an optimized SSE my development, C++ library for face detection, you can immediately use it for your video surveillance system. Face article classifier training data with my webcam images, which are collected at different times, in different lighting, different background environment, it is almost real-time detection of me (a :). Training of non-face data collection different backgrounds, with the same webcam. Was extracted face region, has been subjected to the following treatment: Gaussi Platform: |
Size: 15348736 |
Author:周文活 |
Hits:
Description: 《图像处理的偏微分方程方法》随书光盘完整版
本光碟中包含五个文件夹。
（1）视频剪辑：可供教学演示，其中， two_cells 是采用改进的变分水平集方法，实现GAC模型图像分割的演化过程； denoissing 是利用P_M方程，对图像平滑去噪的演化过程 curve_linear_heat_flow 是利用FT实现的闭合曲线的线性热流演化过程。
（2）二值图像：其中的图像可供形态学图像处理实验用，也可通过提取对象的边界，供曲线演化实验使用。
（3）灰度图象和彩色图像：其中的图像，可以作为图像分割，平滑滤波，反差增强，彩色增强以及图像放大等实验的素材。
（4) MATLAB程序：其中包含10余个MATLAB程序或（函数）的源代码。程序中所作的注释，与书中关于对应算法的描述是一致的。
本光碟中的所有内容，仅供教学和研究参考。-" Image Processing Based on PDE" CD with the book contains the full version of the disc five folders. (1) Video clip: for teaching demonstration, which, " two_cells" is improved variational level set methods to achieve the evolution of GAC model image segmentation " denoissing" is the use of P_M equation, the evolution of image smoothing denoising process " curve_linear_heat_flow" is the use of a closed curve FT achieve a linear heat evolution. (2) binary image: one of the images for morphological image processing experiments, it can also be obtained by extracting the object boundary curve evolution for experimental use. (3) gray-scale images and color images: images which can be used as image segmentation, filtering, contrast enhancement, color enhancement and image amplification test material. (4) MATLAB Program: which includes more than 10 or MATLAB program (function) of the source code. Comments made in the program, and a description of the boo Platform: |
Size: 7414784 |
Author:marshal |
Hits:
Description: This toolbox is meant to facilitate the manipulation of images and video in Matlab. Its purpose is to complement, not replace, Matlab's Image Processing Toolbox, and in fact it requires that the Matlab Image Toolbox be installed. Emphasis has been placed on code efficiency and code reuse. Thanks to everyone who has given me feedback - you've helped make this toolbox more useful and easier to use.(The toolbox is divided into 7 parts, arranged by directory:
channels Robust image features, including HOG, for fast object detection.
classify Fast clustering, random ferns, RBF functions, PCA, etc.
detector Aggregate Channel Features (ACF) object detection code.
filters Routines for filtering images.
images Routines for manipulating and displaying images.
matlab General Matlab functions that should have been a part of Matlab.
videos Routines for annotating and displaying videos.) Platform: |
Size: 9680896 |
Author:redkisses
|
Hits:
Description: 在这个练习中，您将实现线性回归，并看到它在数据上工作。在开始这个编程练习之前，我们强烈建议观看视频讲座，并完成相关主题的复习题。开始锻炼，你将需要下载的启动代码，解压其内容目录到你希望完成练习。如果需要的话，在开始练习之前，使用八度/ matlab中的cd命令改变这个目录。你也可以?ND指示在“环境设置说明”的课程网站安装倍频/ MATLAB。(In this exercise, you will implement linear regression and get to see it work on data. Before starting on this programming exercise, we strongly recommend watching the video lectures and completing the review questions for the associated topics. To get started with the exercise, you will need to download the starter code and unzip its contents to the directory where you wish to complete the exercise. If needed, use the cd command in Octave/MATLAB to change to this directory before starting this exercise.) Platform: |
Size: 508928 |
Author:hahay23456
|
Hits:
Search - matlab code for video