https://github.com/tikam02/crowdestimation

Crowd Density Estimation Using - DeepLearning
https://github.com/tikam02/crowdestimation

deep-neural-networks

Last synced: about 2 months ago
JSON representation

Crowd Density Estimation Using - DeepLearning

• Host: GitHub
• URL: https://github.com/tikam02/crowdestimation
• Owner: Tikam02
• Created: 2018-07-26T18:45:33.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2020-04-09T10:57:36.000Z (about 5 years ago)
• Last Synced: 2025-03-16T22:11:49.229Z (about 2 months ago)
• Topics: deep-neural-networks
• Language: Jupyter Notebook
• Homepage:
• Size: 45.9 MB
• Stars: 31
• Watchers: 3
• Forks: 10
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
Crowd-Density Estimation(Crowd-Count)


Research Paper Implemented



CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly

  Congested Scenes




  


Single-Image Crowd Counting via Multi-Column Convolutional Neural Network




  


CNN-based Cascaded Multi-task Learning of High-level Prior and Density

Estimation for Crowd Counting






About The Project

In this Project my main aim was to learn and implement machine learning

and deep learning in real life.I started from scratch with only knowledge of

python, then i learnt Machine Learning made some programs and then move

forward to deep learning with Crowd-Density Estimation Project.

The project assigned to me was - That is counting the number of peoples in

the crowd and make a ​ Crowd Density Estimation ​ deep learning model to

predict in the real life crowd so that when a threshold of crowd will reach the

application will give alert to that specific area of crowd.

Codes



CCMTL






CSRNet




How to run codes are provided in respective directories of code



Frameworks Used

Deep-learning Frameworks

Tensorflow

Keras

Pytorch

Caffe

Coding Environment and Editor

Jupyter Notebook

Python 3m

Python 2m - CSRNet

GPU

GPU-NVIDIA-950M



Dataset Used and Optimized Predictions Values

Shanghaitech Crowd-Dataset
 

ScreenShot