Our repository is at https://github.com/valdis121/PoVA There are only source files of our latex report without images. We were not able to upload our documentation because of size restrictions. Our had 17Mb and limit in VUT IS is only 10Mb. Here is the link to it - https://www.overleaf.com/8985756448kcmwxkvqjxxy#7e30cd Our prezentation - https://docs.google.com/presentation/d/1GyvNx3bWsOMYYoNqWbjTuuZfVZwnUh0yMxqG3ORn3r8/edit#slide=id.g2a9a0773282_0_29
This project consists of three scripts: geo.py, new.py, and nn.py.
The geo.py script utilizes the Geopy library to create a map and calculate the km/pix distance for a GPS location of a road or crossroad.
The new.py script calculates the homographic matrix and stores it into h.npy. This matrix is used to transform frames in the video to a view similar to an image from Geopy.
The nn.py script uses the YOLOv8 model to estimate the speed of cars in a video. It utilizes the homographic matrix from h.npy to transform each frame and create a view similar to an image from Geopy.
Please refer to the individual script files for more detailed information on their usage and dependencies.
To run this project, you will need to install the following dependencies:
- Geopy
- ultralytics
You can install these dependencies using pip:
pip install geopy
pip install ultralytics