uestcchicken / exMCNN

Extended Multi-Column Convolutional Neural Network for Crowd Counting

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Extended Multi-Column Convolutional Neural Network for Crowd Counting

Implementation of PCM 2018 paper
Extended Multi-Column Convolutional Neural Network for Crowd Counting
using tensorflow

installation

  1. Install tensorflow
  2. git clone
  3. download VGG16 model file (vgg16.npy) from https://github.com/machrisaa/tensorflow-vgg

data setup

All the data setup process follows the pytorch version implementation:
svishwa/crowdcount-mcnn

train

tensorflow:
run python3 train.py A(or B)
model is saved to modelA/ or modelB/

test

run python3 test.py A(or B)

(uncomment code containing heatmap in network.py to generate heatmap)

heatmap

actual: 1110

predicted: 1181

About

Extended Multi-Column Convolutional Neural Network for Crowd Counting


Languages

Language:Python 99.9%Language:Shell 0.1%