使用環境說明:Win10+VS2013+MATLAB R2016a
1,dollar工具箱下載編譯
https://pdollar.github.io/toolbox/,最新版本是Version3.5
For version history click here. This code is licensed under the Simplified BSD License.
- Recommended: download the latest version from GitHub
- Version 3.50 May 12, 2016
- Version 3.00 Aug. 06, 2012
- Version 2.00 Sep. 30, 2007
- Version 1.03 May. 03, 2006 [compatible with cuboids code]
- Version 1.00 Oct. 02, 2005
下載後添加目錄、編譯即可使用
addpath(genpath('ROOT\piotr_toolbox\toolbox\'))
savepath
toolboxCompile
注意編譯前需要配置好VS編譯器
工具箱包含七大部分,channels、classify、detector、filter、images、matlab、videos,後續會對每部分做詳細的說明介紹
2,測試已訓練模型
可以準備好Inria或者Caltech數據使用detector中acfDemoInria.m或者acfDemoCal.m測試,也可以直接用圖片測試
load('detector/models/AcfInriaDetector.mat')
加載的detector分爲三部分:opts、clf、info
opts保存模型相關信息
clf保存分類器信息
info保存特徵通道信息
image = imread('test.jpg')
box = acfDetect(image, detector)
figure;
im(image);
bbApply('draw', box);
% The detector was introduced and described through the following papers:
% [1] P. Doll醨, Z. Tu, P. Perona and S. Belongie
% "Integral Channel Features", BMVC 2009.
% [2] P. Doll醨, S. Belongie and P. Perona
% "The Fastest Pedestrian Detector in the West," BMVC 2010.
% [3] P. Doll醨, R. Appel and W. Kienzle
% "Crosstalk Cascades for Frame-Rate Pedestrian Detection," ECCV 2012.
% [4] P. Doll醨, R. Appel, S. Belongie and P. Perona
% "Fast Feature Pyramids for Object Detection," PAMI 2014.
% [5] W. Nam, P. Doll醨, and J.H. Han
% "Local Decorrelation For Improved Pedestrian Detection," NIPS 2014.
% Please see: http://vision.ucsd.edu/~pdollar/research.html#ObjectDetection