/Real-time-Traffic-and-Pedestrian-Counting

Real-time Traffic and Pedestrian Counting (YOLOV3 in tensorflow2)

Primary LanguagePythonMIT LicenseMIT

Real-time-Traffic-and-Pedestrian-Counting

Introduction

This project focuses " counting and statistics of moving targets we care about ", drive by YOLOv3 which was Implemented in Tensorflow2."
It needs to be stated that the YOLOv3 detector of this project is forked from the nice implementation of YunYang1994

Project Demo

  • The demo is available on Youtube and Bilibili
  • on my laptop gtx1060 FPS reached 12-20

car person

Installation

Reproduce the environment

 conda env create -f environment.yml
 wget https://pjreddie.com/media/files/yolov3.weights

two test videos are prepared here, you should download.

Parameter adjustment

  • For video_demo.py
    •  video_path = "./vehicle.mp4"
    • num_classes = 80
    • utils.load_weights(model, "./yolov3.weights")
  • For utils.py
    • specified_class_id_filter = 2
    • line = [(0, 530), (2100, 530)]

Run demo:

conda activate your_env_name
python video_demo.py

Citation

If you use this code for your publications, please cite it as:

@ONLINE{vdtc,
    author = "Clemente420",
    title  = "Real-time-Traffic-and-Pedestrian-Counting",
    year   = "2020",
    url    = "https://github.com/Clemente420/Real-time-Traffic-and-Pedestrian-Counting"
}

Author

License

This system is available under the MIT license. See the LICENSE file for more info.