CIFAR-YOLO-Vision

This repository contains code for image classification using a Convolutional Neural Network (CNN) on the CIFAR-10 dataset, as well as object detection using the YOLO (You Only Look Once) algorithm.

The object detection model is built using TensorFlow and Keras, while the YOLO algorithm is implemented using OpenCV and Darknet. The repository provides a comprehensive example of how to train a CNN for image classification and perform object detection on images. The code includes data preprocessing, model training, and visualization of results. It serves as a practical guide for understanding and implementing these computer vision techniques.

Object Detection

The code demonstrates object detection using the YOLO algorithm:

1. Loading and preprocessing the CIFAR-10 dataset.
2. Building a CNN model for Object Detection.
3. Training the model on the dataset.
4. Downloading and Loading the YOLO model files and class names.
5. Evaluating the model's performance on test data.

Usage

To use the code in this repository, follow these steps:

1. Install the required dependencies: TensorFlow, Keras, OpenCV.
2. Clone the repository: git clone https://github.com/theakshat1/CIFAR-YOLO-Vision
3. Navigate to the repository directory: cd CIFAR-YOLO-Vision
4. Install the required dependencies using pip: pip install -U -r requirements.txt

Modify YOLOv3.ipynb according to need. (Project made on Google Colab)

Resources

• CIFAR-10 Dataset
• YOLO (You Only Look Once) Algorithm

License

This project is licensed under the MIT License. See the LICENSE file for details.