spatial-detector 0.1.0

3D object detection and spatial mapping using webcam

Spatial Detector

A Python package for 3D object detection and spatial mapping using a webcam.

Features

• YOLO-based object detection
• MiDaS depth estimation
• 3D spatial mapping and localization
• Top-down view visualization
• Depth calibration for accurate measurements
• Designed for Apple Silicon and CUDA GPUs

Installation

pip install spatial-detector

Or install from source:

git clone https://github.com/yourusername/spatial-detector.git
cd spatial-detector
pip install -e .

Usage

Command Line Interface

# Basic usage
spatial-detector

# With custom settings
spatial-detector --camera 0 --yolo-model yolov8n.pt --device mps

As a Python Library

from spatial_detector.detection import YOLODetector
from spatial_detector.depth import MiDaSDepthEstimator
from spatial_detector.mapping import SpatialMap, DepthCalibrator

# Initialize components
detector = YOLODetector(model_path="yolov8n.pt")
depth_estimator = MiDaSDepthEstimator()
spatial_map = SpatialMap()

# Process a frame
detections = detector.detect(frame)
depth_map, _ = depth_estimator.estimate_depth(frame)

Controls

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

Calibration

For accurate measurements:

1. Press c to enter calibration mode
2. Place an object at a known distance (e.g., 1 meter from camera)
3. Use +/- to set the correct distance
4. Align the object with the center crosshair
5. Press space to calibrate
6. Press s to save the calibration

License

MIT