This repository contains Python code for binary image classification using Convolutional Neural Networks (CNNs).
In the 15frames file we pull every 15th frame from the video sequence. Then we manually or with the help of yolov5 divide all frames into different folders. The first folder is the person sitting at the table. The second folder - no one is at the table. Machine learning. I used pre-trained models, ResNet18 and InceptionV3. Worked with both TensorFlow and PyTorch.
The following packages are required to run the code:
- Python (>= 3.6)
- NumPy
- TensorFlow (>= 2.0)
- Keras (>= 2.0)
- PyTorch (>= 1.0)
https://drive.google.com/file/d/1yhNKlZ52q6XlnZZe8KIQPYgFGiV5LkPA/view?usp=sharing - link to download dataset. I manually selected 4k+ images each for two different folders.
https://disk.yandex.ru/i/9AuLkzsd55083w - link to download video
This project was inspired by the many great examples and tutorials available online on image classification with CNNs. Special thanks to the TensorFlow and Pytorch teams for providing excellent libraries for deep learning in Python.
This project is licensed under the MIT License
