Computer Vision 101 Workshop

A practical workshop on object detection using YOLOv8 and computer vision fundamentals.

📌 Overview

This repository contains resources and code for a hands-on workshop introducing computer vision and object detection. Participants will learn how to:

✅ Set up their development environment
✅ Train custom YOLOv8 models
✅ Implement real-time object detection using webcams

---

🚀 Quick Links

🔗 Google Colab Notebook - Model Training
📁 Workshop Files Google Drive
📸 Upload Images for Training

---

📋 Requirements

• Python 3.11 or later
• Webcam (for real-time detection)
• GitHub account (optional)
• Google account (for Google Colab)

---

🛠️ Installation Guide

For macOS

1️⃣ Install Anaconda

chmod +x Anaconda3-2024.10-1-MacOSX-arm64.sh
./Anaconda3-2024.10-1-MacOSX-arm64.sh
conda --version

2️⃣ Create & Activate Environment

conda create --name yolov8-env python=3.11 -y
conda activate yolov8-env

3️⃣ Install Required Packages

conda update conda
conda install -c pytorch pytorch torchvision
pip install ultralytics
pip install label-studio

4️⃣ Start Label Studio

label-studio start

📌 Open http://localhost:8080 in your browser.

---

For Windows

1️⃣ Install Anaconda

• Download from Anaconda Download Page
• Follow installation instructions
• Verify installation:

  conda --version

2️⃣ Create & Activate Environment

conda create --name yolov8-env python=3.11 -y
conda activate yolov8-env

3️⃣ Install Required Packages

conda update conda
conda install -c pytorch pytorch torchvision
pip install ultralytics
pip install label-studio

4️⃣ Start Label Studio

label-studio start

📌 Access at http://localhost:8080.

---

📖 Workshop Workflow

1️⃣ Setup Environment

• Install dependencies (as described above)
• Clone this repository:

  git clone https://github.com/YOUR_USERNAME/computer-vision-101.git
  cd computer-vision-101

2️⃣ Prepare Dataset

• Collect images for custom object detection
• Upload to the Google Drive folder
• Annotate using Label Studio or CVAT

3️⃣ Train Model

• Use the Google Colab notebook
• Upload your annotated dataset
• Train the YOLOv8 model
• Download the trained model weights

4️⃣ Run Inference

• Download the trained model from Colab
• Run the provided object detection script
• Modify detection parameters as needed

---

💻 Example Usage

1️⃣ Download your trained model from Google Colab and place it in your project directory.
2️⃣ Modify the model_path variable in the provided script to point to your model file.
3️⃣ Run the detection script:

python object_detection.py

4️⃣ Controls:

• Press q to quit
• Press s to save a screenshot

---

🛠️ Resources & Tools

🔍 Labeling/Annotation Software

• CVAT
• Label Studio

📂 Image Datasets

• Open Images Dataset
• COCO Dataset

🛠️ Libraries & Frameworks

• Ultralytics YOLOv8
• PyTorch
• Anaconda

---

🔧 Troubleshooting

⚠️ Permission Errors

• Run your terminal as Administrator (Windows) or use sudo (macOS).

⚠️ Environment Issues

• Ensure your environment is activated:

  conda activate yolov8-env

⚠️ Dependency Conflicts

• Regularly update pip and conda:

  pip install --upgrade pip
  conda update conda

⚠️ Webcam Issues

• If your webcam isn't detected, change the device index:

  cv2.VideoCapture(0)  # Try changing 0 to 1 or 2

---

📜 License

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

---

🙌 Acknowledgments

🙏 Special thanks to:

• Ultralytics for YOLOv8
• All workshop participants for their contributions

---