YoloZone

YoloZone is a powerful computer vision toolkit built on YOLOv8, providing intuitive interfaces for object detection, pose estimation, and object tracking. It simplifies complex computer vision tasks with easy-to-use APIs and comprehensive documentation.

Features

Object Detection

• Detect and classify objects in images and videos
• Support for custom models and multiple detection strategies
• Real-time processing capabilities
• Configurable confidence thresholds

Pose Estimation

• Advanced human pose detection and keypoint analysis
• Real-time pose tracking
• 17-point keypoint detection
• Angle and distance measurements between keypoints
• Support for multiple people in frame

Object Tracking

• Robust object tracking across video frames
• Motion pattern analysis
• Trajectory data generation
• Line crossing detection
• Multi-object tracking

Installation

pip install yolozone

Quick Start

Object Detection

from yolozone import Objects

# Initialize detector
detector = Objects()

# Detect objects in an image
results = detector.detect('image.jpg')

# Process video stream
detector.process_video('video.mp4', output='output.mp4')

Pose Estimation

from yolozone import Pose

# Initialize pose estimator
pose = Pose()

# Detect poses in an image
results = pose.detect('image.jpg')

# Get keypoints
for detection in results:
    keypoints = pose.get_keypoints(detection)
    print(f"Found person with {len(keypoints)} keypoints")

Object Tracking

from yolozone import Tracker

# Initialize tracker
tracker = Tracker()

# Track objects in video
tracks = tracker.track_video('video.mp4')

# Analyze motion patterns
for track in tracks:
    motion = tracker.analyze_motion(track)
    print(f"Track {track.id}: {motion.pattern}")

Keypoint Reference

The pose estimation module uses the following 17 keypoints:

ID	Keypoint	ID	Keypoint
0	Nose	9	Left Wrist
1	Left Eye	10	Right Wrist
2	Right Eye	11	Left Hip
3	Left Ear	12	Right Hip
4	Right Ear	13	Left Knee
5	Left Shoulder	14	Right Knee
6	Right Shoulder	15	Left Ankle
7	Left Elbow	16	Right Ankle
8	Right Elbow

Documentation

Visit our comprehensive documentation for:

• Detailed API references
• Code examples
• Implementation guides
• Best practices
• Troubleshooting tips

Requirements

• Python 3.7+
• ultralytics (YOLOv8)
• opencv-python
• numpy

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

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

Author

• Nushan Kodikara
  • GitHub: @nushankodikara
  • LinkedIn: Nushan Kodikara
  • Email: [email protected]

References

• Ultralytics YOLOv8
• YOLOv8 Pose Estimation