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