基于条件对抗生成神经网络的ACSCP人群密度估计模型----ACSCP crowd counting model

github

ACSCP crowd counting model

=======

Introduction

This is open source project for crowd counting. Implement with paper “Crowd Counting via Adversarial Cross-Scale Consistency Pursuit” from Shanghai Jiao Tong University. For more details, please refer to our Baidu Yun

Contents

1. Installation
2. Preparation
3. Train/Eval/Release
4. Additional
5. Details

Installation

1 Configuration requirements

python3.x

Please using GPU, suggestion more than GTX960

python-opencv
#tensorflow-gpu==1.0.0
#tensorflow==1.0.0
scipy==1.0.1
matplotlib==2.2.2
numpy==1.14.2

conda install -c https://conda.binstar.org/menpo opencv3
pip install -r requirements.txt

2 Get the code

git clone https://github.com/Ling-Bao/ACSCP_cGAN
cd ACSCP_cGAN

### Preparation 1 ShanghaiTech Dataset. ShanghaiTech Dataset makes by Zhang Y, Zhou D, Chen S, et al. For more detail, please refer to paper “Single-Image Crowd Counting via Multi-Column Convolutional Neural Network” and click on [here](https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Zhang_Single-Image_Crowd_Counting_CVPR_2016_paper.pdf). 2 Get dataset and its corresponding map label [Baidu Yun](https://pan.baidu.com/s/1gccvnvIeLgQZCVuA6iZEjA) Password: yvs1 3 Unzip dataset to ACSCP_cGAN root directory

unzip Data.zip

### Train/Eval/Release Train is easy, just using following step. 1 Train. Using [main.py](main.py) to evalute crowd counting model

python main.py --phase train

2 Eval. Using [main.py](main.py) to evalute crowd counting model

python main.py --phase test

OR

python main.py --phase inference

3 Model release Model release. Using [product.py](product.py) to release crowd counting model. Download release version 0.1.0, please click on [here](release/version1.0.0.tar.gz) ### Addtional 1 Crowd map generation tools Source code store in “data_maker”, detail please check [here](data_maker/README.md). **Note:** This tools write by matlab, please install matlab. 2 Results

Original image



Real crowd map, counting is 707



Predict crowd map, counting is 698

3 crowd counting paper collection, thanks for **gjy3035** **Github:**[ Awesome-Crowd-Counting](https://github.com/gjy3035/Awesome-Crowd-Counting) **Density Map Generation from Key Points:** [[Matlab Code]](https://github.com/aachenhang/crowdcount-mcnn/tree/master/data_preparation) [[Python Code]](https://github.com/leeyeehoo/CSRNet-pytorch/blob/master/make_dataset.ipynb) ### Details 1 Tring to delete dropout layers. 2 Improving activation funtion for last layer to adapt crowd counting map estimation.

=======