资源简介
基于Fisher Vector的人脸检测 ,内用vlfeat库,附详细的代码
代码片段和文件信息
% Copyright (c) 2014 Karen Simonyan
% All rights reserved.
% This code is made available under the terms of the BSD license (see COPYING file).
clear;
% set to ‘false‘ to run locally
isCluster = false;
% experiment name
prms.expName = ‘SIFT_1pix_PCA64_GMM512‘;
% image set name
prms.setName = ‘lfw_aligned‘;
% ‘unrest‘: unrestricted ‘rest‘: restricted
prms.trainSettingName = ‘unrest‘;
% prms.expName = ‘SIFT_1pix_PCA64_GMM512_restricted‘;
% prms.setName = ‘lfw_funneled‘;
% prms.trainSettingName = ‘rest‘;
%%
if isCluster
% run on the cluster
fprintf(‘Computing DSIFT...\n‘);
JD1 = batch(‘face_desc.manager.face_descriptor.compute_dense‘ ‘matlabpool‘ 50 ‘workspace‘ prms);
wait(JD1);
fprintf(‘Learning PCA and codebooks...\n‘);
JD2 = batch(‘face_desc.manager.face_descriptor.learn_pca_gmm‘ ‘matlabpool‘ 10 ‘workspace‘ prms);
wait(JD2);
fprintf(‘Computing FV...\n‘);
JD3 = batch(‘face_desc.manager.face_descriptor.compute_fv‘ ‘matlabpool‘ 50 ‘workspace‘ prms);
else
% copy params struct to the current workspace & run the code
prmsName = fieldnames(prms);
for idxName = 1:numel(prmsName)
assignin(‘base‘ prmsName{idxName} prms.(prmsName{idxName}));
end
% subset of dense SIFT features
face_desc.manager.face_descriptor.compute_dense;
% SIFT-PCA & GMM learning
face_desc.manager.face_descriptor.learn_pca_gmm;
% FV computation
face_desc.manager.face_descriptor.compute_fv;
end
- 上一篇:OPCClient.exe
- 下一篇:CAD窗口并排显示
相关资源
- Vectorcontrolinductionmotor.rar
- Nonlinear Separation and Optimality Conditions
- The Inner Workings - of - word2vec :一文搞懂
- Finite-Dimensional Vector Spaces - P. Halmos (
- WEEE 垃圾桶标志 矢量图
- VECTOR公司的AUTOSAR入门资料,非常经典
- LitJson.rar
-
XL Driver Library - Desc
ription - canoe 11.0 安装软件,基于硬件的licen
- GoogleNews-vectors-negative300.bin.gz
- A tutorial on support vector machines for patt
- A_Tutorial_on_Support_Vector_Machines_for_Patt
- VS2015调试vector的文件
- CAD图形引擎库VectorDraw6.0web
- HexViewVectorV1.09.01
- fisher判别法
- vector_map.zip
评论
共有 条评论