C-COT跟踪算法在OTB数据集上测试的接口函数

C-COT算法作者其实已经给出了在几个数据集中测试的原始结果，其中也包括了OTB数据集。但是如果我们想要自己测试需要怎么写接口函数呢？

很简单，把代码中的 setup_paths.m 和在runfiles文件夹中的 test.m 文件中的内容综合一下就行了，这里需要把整个文件夹的名字从 C-COT 改为 CCOT，因为命名里面短横线不符合matlab文件命名规范。然后在CCOT文件夹下新建run_CCOT.m 文件。
run_CCOT.m

function results=run_CCOT(seq, res_path, bSaveImage)close all%% Add the neccesary paths
[pathstr, name, ext] = fileparts(mfilename('fullpath'));
% Tracker implementation
addpath([pathstr '/implementation/']);
% Runfiles
addpath([pathstr '/runfiles/']);
% Utilities
addpath([pathstr '/utils/']);
% The feature extraction
addpath(genpath([pathstr '/feature_extraction/']));
% Matconvnet
addpath([pathstr '/external_libs/matconvnet/matlab/mex/']);
addpath([pathstr '/external_libs/matconvnet/matlab']);
addpath([pathstr '/external_libs/matconvnet/matlab/simplenn']);
% PDollar toolbox
addpath(genpath([pathstr '/external_libs/pdollar_toolbox/channels']));
% Mtimesx
addpath([pathstr '/external_libs/mtimesx/']);
%%% Feature specific parameters
hog_params.cell_size = 4;grayscale_params.colorspace='gray';
grayscale_params.cell_size = 1;cn_params.tablename = 'CNnorm';
cn_params.useForGray = false;
cn_params.cell_size = 4;ic_params.tablename = 'intensityChannelNorm6';
ic_params.useForColor = false;
ic_params.cell_size = 4;cnn_params.nn_name = 'imagenet-vgg-m-2048.mat'; % Name of the network
cnn_params.output_layer = [0 3 14];             % Which layers to use
cnn_params.downsample_factor = [4 2 1];         % How much to downsample each output layer
cnn_params.input_size_mode = 'adaptive';        % How to choose the sample size
cnn_params.input_size_scale = 1;                % Extra scale factor of the input samples to the network (1 is no scaling)% Which features to include
params.t_features = {struct('getFeature',@get_cnn_layers, 'fparams',cnn_params),......struct('getFeature',@get_colorspace, 'fparams',grayscale_params),......struct('getFeature',@get_fhog,'fparams',hog_params),......struct('getFeature',@get_table_feature, 'fparams',cn_params),......struct('getFeature',@get_table_feature, 'fparams',ic_params),...
};% Global feature parameters
params.t_global.normalize_power = 2;    % Lp normalization with this p
params.t_global.normalize_size = true;  % Also normalize with respect to the spatial size of the feature
params.t_global.normalize_dim = true;   % Also normalize with respect to the dimensionality of the feature% Image sample parameters
params.search_area_shape = 'square';    % The shape of the samples
params.search_area_scale = 5.0;         % The scaling of the target size to get the search area
params.min_image_sample_size = 200^2;   % Minimum area of image samples
params.max_image_sample_size = 300^2;   % Maximum area of image samples% Detection parameters
params.refinement_iterations = 1;       % Number of iterations used to refine the resulting position in a frame
params.newton_iterations = 5;           % The number of Newton iterations used for optimizing the detection score% Learning parameters
params.output_sigma_factor = 1/12;     % Label function sigma
params.learning_rate = 0.0075;         % Learning rate
params.nSamples = 400;                  % Maximum number of stored training samples
params.sample_replace_strategy = 'lowest_prior';    % Which sample to replace when the memory is full
params.lt_size = 0;                     % The size of the long-term memory (where all samples have equal weight)% Conjugate Gradient parameters
params.max_CG_iter = 5;                 % The number of Conjugate Gradient iterations
params.init_max_CG_iter = 100;          % The number of Conjugate Gradient iterations used in the first frame
params.CG_tol = 1e-3;                   % The tolerence of CG does not have any effect
params.CG_forgetting_rate = 10;         % Forgetting rate of the last conjugate direction
params.precond_data_param = 0.5;        % Weight of the data term in the preconditioner
params.precond_reg_param = 0.01;        % Weight of the regularization term in the preconditioner% Regularization window parameters
params.use_reg_window = true;           % Use spatial regularization or not
params.reg_window_min = 1e-4;          % The minimum value of the regularization window
params.reg_window_edge = 10e-3;         % The impact of the spatial regularization
params.reg_window_power = 2;            % The degree of the polynomial to use (e.g. 2 is a quadratic window)
params.reg_sparsity_threshold = 0.05;   % A relative threshold of which DFT coefficients that should be set to zero% Interpolation parameters
params.interpolation_method = 'bicubic';    % The kind of interpolation kernel
params.interpolation_bicubic_a = -0.75;     % The parameter for the bicubic interpolation kernel
params.interpolation_centering = true;      % Center the kernel at the feature sample
params.interpolation_windowing = false;     % Do additional windowing on the Fourier coefficients of the kernel% Scale parameters
params.number_of_scales = 5;            % Number of scales to run the detector
params.scale_step = 1.02;               % The scale factor% Other parameters
params.visualization = 0;               % Don't visualiza tracking and detection scores
params.debug = 0;                       % Do full debug visualizations_frames = seq.s_frames;% Initialize
params.wsize = [seq.init_rect(1,4), seq.init_rect(1,3)];
params.init_pos = [seq.init_rect(1,2), seq.init_rect(1,1)] + floor(params.wsize/2);
params.s_frames = s_frames;% Run tracker
results = tracker(params);if bSaveImageimwrite(frame2im(getframe(gcf)),[res_path num2str(frame) '.jpg']); endend

C-COT跟踪算法在OTB数据集上测试的接口函数相关推荐

HCFT和HCFTstar在OTB数据集中测试的接口函数
HCFT跟踪算法(Hierarchical Convolutional Features for Visual Tracking)是2015年发表在ICCV上的一篇结合相关滤波和深度特征的跟踪算法,作 ...
什么是维度诅咒？如何评估降维算法在当前任务数据集上的效果？
什么是维度诅咒?如何评估降维算法在当前任务数据集上的效果? 什么是维度诅咒? 随着维数(或特征)个数的增加,任意两个样本或者实例之间的平均距离也会增加,换句话说,在高维空间中,实例往往位于空间的边缘. ...
感知机算法在鸢尾花数据集上的实践
感知机算法在鸢尾花数据集上的实践 1 问题背景 2 数据集的微处理 3 导数据,并进行简单可视化 4 模型算法的训练部分 1 问题背景感知机作为一个比较经典的处理线性二分类的算法,今天想着实践一下, ...
MMDetection（三）：公开数据集上测试和训练模型
MMDetection(三):公开数据集上测试和训练模型 1. 准备数据集 2. 测试数据集 2.1 举例 2.2 网络摄像头演示 2.3 测试图像和视频的同步接口 3. 训练模型 3.1 使用单个G ...
用DPM(Deformable Part Model，voc-release3.1)算法在INRIA数据集上训练自己的人体检测模型
我的环境 DPM源码版本:voc-release3.1 VOC开发包版本:VOC2007_devkit_08-Jun Matlab版本:MatlabR2012b c++编译器:VS2010 系统:Wi ...
几种聚类算法在IRIS数据集上的应用（python）
分解聚类.C均值聚类及其改进.ISODATA应用在IRIS数据集的表现 IRIS数据集分解聚类 C均值聚类 ISODATA K均值聚类这个其实是之前的一次课程作业,要求是对鸢尾花数据集IRIS做聚 ...
Pytorch 实现全连接神经网络/卷积神经网络训练MNIST数据集，并将训练好的模型在制作自己的手写图片数据集上测试
使用教程代码下载地址:点我下载模型在训练过程中会自动显示训练进度,如果您的pytorch是CPU版本的,代码会自动选择CPU训练,如果有cuda,则会选择GPU训练. 项目目录说明: CNN文件夹 ...
Python机器学习：KNN算法03训练数据集，测试数据集train test split
示例代码首先引入相关包 import numpy as np import matplotlib.pyplot as plt from sklearn import datasets import ...
CVPR2019 oral 目标跟踪算法之SiamRPN++
点上方蓝字计算机视觉联盟获取更多干货在右上方 ··· 设为星标 ★,与你不见不散编辑:Sophia 计算机视觉联盟报道 | 公众号 CVLianMeng 转载于 :lbvic 知乎 http ...
90+目标跟踪算法九大benchmark！基于判别滤波器和孪生网络的视觉目标跟踪：综述与展望...
点击下方卡片,关注"自动驾驶之心"公众号 ADAS巨卷干货,即可获取点击进入→自动驾驶之心[目标跟踪]技术交流群后台回复[视觉目标跟踪综述]获取本文论文! 准确和鲁棒的视觉目标 ...

C-COT跟踪算法在OTB数据集上测试的接口函数

C-COT跟踪算法在OTB数据集上测试的接口函数相关推荐

最新文章

热门文章