// Author: Juergen Gall, BIWI, ETH Zurich
// Email: gall@vision.ee.ethz.ch
// You may use, copy, reproduce, and distribute this Software for any
// non-commercial purpose, subject to the restrictions of the
// Microsoft Research Shared Source license agreement ("MSR-SSLA").
// Some purposes which can be non-commercial are teaching, academic
// research, public demonstrations and personal experimentation. You
// may also distribute this Software with books or other teaching
// materials, or publish the Software on websites, that are intended
// to teach the use of the Software for academic or other
// non-commercial purposes.
// You may not use or distribute this Software or any derivative works
// in any form for commercial purposes. Examples of commercial
// purposes would be running business operations, licensing, leasing,
// or selling the Software, distributing the Software for use with
// commercial products, using the Software in the creation or use of
// commercial products or any other activity which purpose is to
// procure a commercial gain to you or others.
// If the Software includes source code or data, you may create
// derivative works of such portions of the Software and distribute
// the modified Software for non-commercial purposes, as provided
// herein.
// THE SOFTWARE COMES "AS IS", WITH NO WARRANTIES. THIS MEANS NO
// EXPRESS, IMPLIED OR STATUTORY WARRANTY, INCLUDING WITHOUT
// LIMITATION, WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A
// PARTICULAR PURPOSE, ANY WARRANTY AGAINST INTERFERENCE WITH YOUR
// ENJOYMENT OF THE SOFTWARE OR ANY WARRANTY OF TITLE OR
// NON-INFRINGEMENT. THERE IS NO WARRANTY THAT THIS SOFTWARE WILL
// FULFILL ANY OF YOUR PARTICULAR PURPOSES OR NEEDS. ALSO, YOU MUST
// PASS THIS DISCLAIMER ON WHENEVER YOU DISTRIBUTE THE SOFTWARE OR
// DERIVATIVE WORKS.
// NEITHER MICROSOFT NOR ANY CONTRIBUTOR TO THE SOFTWARE WILL BE
// LIABLE FOR ANY DAMAGES RELATED TO THE SOFTWARE OR THIS MSR-SSLA,
// INCLUDING DIRECT, INDIRECT, SPECIAL, CONSEQUENTIAL OR INCIDENTAL
// DAMAGES, TO THE MAXIMUM EXTENT THE LAW PERMITS, NO MATTER WHAT
// LEGAL THEORY IT IS BASED ON. ALSO, YOU MUST PASS THIS LIMITATION OF
// LIABILITY ON WHENEVER YOU DISTRIBUTE THE SOFTWARE OR DERIVATIVE
// WORKS.
// When using this software, please acknowledge the effort that
// went into development by referencing the paper:
//
// Gall J. and Lempitsky V., Class-Specific Hough Forests for
// Object Detection, IEEE Conference on Computer Vision and Pattern
// Recognition (CVPR'09), 2009.
// Note that this is not the original software that was used for
// the paper mentioned above. It is a re-implementation for Linux.
#compile (needs OpenCV)
make all
#clean
make clear
#run
./run.sh mode [config.txt] [tree_offset]
mode: 0 - train; 1 - show leafs; 2 - detect
config.txt: config file
tree_offset: output number for trees (treetable[index+offset].txt)
A config.txt example is given in the subdirectory 'example'
#example train
./run_train.sh
#example detect
./run_detect.sh
Config.txt:
Information for storing and loading trees:
# Path to trees + prefix 'treetable'
/scratch/tmp/forest/example/trees/treetable
# Number of trees
10
Image patch size (needs to be the same for training and learning):
# Patch width
16
# Patch height
16
Information about the test set:
# Path to images
/scratch/tmp/forest/example/testimages
# File with names of images
/scratch/tmp/forest/example/test.txt
Not used:
# Extract features
1
Specify scale and ratio if necessary:
# Scales (Number of scales - Scales)
3 0.5 1 2 // 3 scales with factors 0.5, 1, and 2
# Ratios (Number of ratios - ratio)
1 1 // 1 aspect ratio with factor 1 = fixed aspect ratio
Information for storing the Hough images (scaled by a fixed factor such that max<=255):
# Output path
/scratch/tmp/forest/example/detect
# Scale factor for output image (default: 128)
50
Information about positive and negative examples:
# Path to positive examples
/scratch/tmp/forest/example/trainimages
# File with positive examples
/scratch/tmp/forest/example/train_pos.txt
# Subset of positive images -1: all images
-1
# Sample patches from pos. examples
50
# Path to negative examples
/scratch/tmp/forest/example/trainimages
# File with negative examples
/scratch/tmp/forest/example/train_neg.txt
# Subset of negative images -1: all images
-1
# Sample patches from neg. examples
50
train_neg.txt:
50 1 // number of images + dummy value (1)
neg0.png 0 0 100 40 // filename + boundingbox (top left - bottom right)
train_pos.txt:
50 1 // number of images + dummy value (1)
pos0.png 0 0 74 36 37 18 // filename + boundingbox (top left - bottom right) + center of bounding box
Output:
detect-[I]_sc[S]_c[R].png
I: image id
S: scale id
C: ratio id
The images are slides of the 4D (x,y,scale,ratio) voting space for an image I.
For detection, the space needs to be smoothed and the maxima detected (or using mean shift).
This is not part of the source code.
Juergen Gall
BIWI, ETH Zurich 17 July 2009
没有合适的资源?快使用搜索试试~ 我知道了~
温馨提示
用于目标检测的Hough森林分类法。比随即森林效果要好。附有文章和代码。代码是在unix环境下的Opencv运行的。还有训练数据。
资源推荐
资源详情
资源评论
收起资源包目录
class-specific-Hough-forests-.rar_Forests_Hough forests_Linux的op (258个子文件)
CRForest-Detector.cpp 16KB
CRPatch.cpp 14KB
CRTree.cpp 13KB
CRForestDetector.cpp 4KB
HoG.cpp 2KB
forest.tar.gz 6.99MB
CRTree.h 5KB
CRForest.h 2KB
CRPatch.h 2KB
HoG.h 1KB
CRForestDetector.h 685B
tudROC.m 638B
horseROC.m 559B
compareDET.m 446B
plotHOG.m 386B
plotDET.m 310B
Makefile 552B
4DISM-train210_roc.mat 4KB
HOG_roc.mat 2KB
Art-ISM-train400_roc.mat 2KB
roc_mr.mat 594B
roc2.mat 565B
roc_sc2_b.mat 556B
roc_sc2_a.mat 554B
roc_mr.mat 372B
Class-Specific Hough Forests for Object Detection.pdf 4.66MB
detect-14_sc2_c0.png 5KB
detect-12_sc2_c0.png 5KB
detect-15_sc2_c0.png 5KB
detect-8_sc2_c0.png 5KB
detect-17_sc2_c0.png 5KB
detect-16_sc2_c0.png 5KB
detect-18_sc2_c0.png 5KB
detect-10_sc2_c0.png 5KB
detect-11_sc2_c0.png 5KB
detect-5_sc2_c0.png 5KB
detect-6_sc2_c0.png 5KB
detect-1_sc2_c0.png 5KB
detect-2_sc2_c0.png 5KB
detect-4_sc2_c0.png 5KB
detect-9_sc2_c0.png 4KB
detect-13_sc2_c0.png 4KB
detect-7_sc2_c0.png 4KB
detect-3_sc2_c0.png 4KB
detect-0_sc2_c0.png 4KB
pos21.png 4KB
neg42.png 4KB
neg37.png 4KB
neg4.png 4KB
neg1.png 4KB
neg12.png 4KB
neg3.png 4KB
neg25.png 4KB
neg2.png 4KB
neg20.png 4KB
neg10.png 4KB
neg35.png 4KB
neg13.png 4KB
neg16.png 4KB
neg26.png 4KB
neg39.png 4KB
neg21.png 4KB
neg29.png 4KB
neg41.png 4KB
neg7.png 4KB
neg8.png 4KB
neg15.png 4KB
neg48.png 4KB
neg19.png 4KB
neg43.png 4KB
neg11.png 4KB
neg47.png 3KB
pos17.png 3KB
neg5.png 3KB
neg0.png 3KB
neg24.png 3KB
test8.png 3KB
neg49.png 3KB
neg17.png 3KB
neg6.png 3KB
neg14.png 3KB
neg28.png 3KB
neg32.png 3KB
neg44.png 3KB
pos16.png 3KB
neg31.png 3KB
neg33.png 3KB
neg45.png 3KB
neg18.png 3KB
neg46.png 3KB
neg22.png 3KB
test5.png 3KB
test6.png 3KB
pos10.png 3KB
neg34.png 3KB
pos24.png 3KB
test7.png 3KB
neg38.png 3KB
test1.png 3KB
neg27.png 3KB
共 258 条
- 1
- 2
- 3
资源评论
邓凌佳
- 粉丝: 65
- 资源: 1万+
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功