v0.1.0

Spatial Detector v0.1.0

Release Notes - Spatial Detector v0.1.0

We're excited to announce the first release of Spatial Detector, a Python package for 3D object detection and spatial mapping using a webcam!

🌟 Highlights

• Object Detection: Identify objects in real-time using YOLOv8
• Depth Estimation: Calculate accurate distances using monocular depth sensing
• Spatial Mapping: Create top-down maps of your environment
• Apple Silicon Support: Optimized for M1/M2 Macs using Metal Performance Shaders
• Interactive Visualization: See depth maps and object positions in real-time

🛠️ Features

Core Capabilities

• Real-time object detection and classification
• Monocular depth estimation without specialized hardware
• 3D localization of detected objects
• Persistence and tracking of objects across frames
• Top-down spatial mapping visualization

Usability

• Simple command-line interface
• Interactive calibration for accurate measurements
• Visualization controls (toggle depth, labels, etc.)
• Video recording capabilities
• Real-time FPS counter

Technical

• Optimized for Apple Silicon and NVIDIA GPUs
• UV package manager integration
• Modern Python packaging with pyproject.toml
• GitHub Actions for CI/CD

📋 Installation & Usage

Installation

uv pip install spatial-detector

Basic Usage

spatial-detector

With Options

spatial-detector --device mps --yolo-model yolov8s.pt --width 1280 --height 720

🔍 Controls

• q: Quit
• d: Toggle depth visualization
• l: Toggle labels
• c: Enter calibration mode
• +/-: Adjust calibration distance
• space: Set calibration point
• s: Save depth calibration

🔮 What's Next

In upcoming releases, we plan to add:

• Multi-camera support
• 3D mesh generation
• SLAM (Simultaneous Localization And Mapping)
• Object segmentation
• More advanced tracking algorithms
• Mobile device support

We welcome contributions and feedback!

What's Changed

Full Changelog: https://github.com/oldhero5/spatial_detector/commits/v0.1.0