No description, website, or topics provided.
Switch branches/tags
Nothing to show
Clone or download
Permalink
Failed to load latest commit information.
Graph add all latent_self_learning May 18, 2018
code3.2.1 add all latent_self_learning May 18, 2018
dpm add all latent_self_learning May 18, 2018
slsvm add all latent_self_learning May 18, 2018
tracker Delete TrackFace.m Jun 3, 2018
Readme.md readme May 18, 2018
Readme.txt change readme May 18, 2018
globals.m add all latent_self_learning May 18, 2018
globals_temporal.m add all latent_self_learning May 18, 2018
learn.exe add all latent_self_learning May 18, 2018
oneClassSVM.m add all latent_self_learning May 18, 2018
proposed-self-learning.png readme May 18, 2018
st_detectBoxesClassify.m add all latent_self_learning May 18, 2018
st_detectBoxesFinal.m add all latent_self_learning May 18, 2018
st_detectBoxesRank.m add all latent_self_learning May 18, 2018
st_detecting.m add all latent_self_learning May 18, 2018
st_dpm_data.m add all latent_self_learning May 18, 2018
st_dpm_detect.m add all latent_self_learning May 18, 2018
st_dpm_detect_img.m add all latent_self_learning May 18, 2018
st_dpm_train.m add all latent_self_learning May 18, 2018
st_edgeBoxesRank.m add all latent_self_learning May 18, 2018
st_fea2scales.m add all latent_self_learning May 18, 2018
st_featBag.m add all latent_self_learning May 18, 2018
st_featpyramid.m add all latent_self_learning May 18, 2018
st_latentSVMLearning.m add all latent_self_learning May 18, 2018
st_learning.m add all latent_self_learning May 18, 2018
st_learningWeight.m add all latent_self_learning May 18, 2018
st_paras_setting.m add all latent_self_learning May 18, 2018
start_up.m add all latent_self_learning May 18, 2018

Readme.md

Self-learning Scene-specific Pedestrian Detectors using a Progressive Latent Model

Introduction

A self-learning approach is proposed towards solving scene-specific pedestrian detection problem without any human¡¯ annotation involved. The selflearning approach is deployed as progressive steps of object discovery, object enforcement, and label propagation.

This is a matlab code of Self-learning Scene-specific Pedestrian Detectors using a Progressive Latent Model. Copyright Reserved by University of Chinese Academy of Sciences. It is free for academy purpose. Please contacet qxye@ucas.ac.cn if you have more problems

Runtime enviroment: Matalb12 or later vergion,

Configuration:

  1. Download the Edgebox proposal generation code from http://vision.ucsd.edu/~pdollar/research.html

  2. Download the DPM code from Ross Grishick's UC berkely websit

  3. Supose the video name is 'PETS09-S2L2.avi', put the video in the dataset 'data'

  4. Make a folder as the name of video

  5. Randomly prepare >1000 negtive images in the data\videoname\neg folder Prepare the neg_filelist.txt in the data\videoname foler

  6. Run Demo by inputting st_learning('.\data\PETS09-S2L2.avi')