Skip to content

Latest commit

 

History

History
 
 

README.md

DeepCamera Source Code

This directory contains the original DeepCamera processing pipeline, built over 4+ years of development. These modules handle everything from camera capture to AI-powered detection and re-identification.

πŸ†• Looking for the new skill system? See skills/ for the pluggable AI skill architecture.

Module Map

Module Purpose Status
yolov7_person_detector/ Real-time person detection using YOLOv7 βœ… Production
yolov7_reid/ Person re-identification β€” tracks individuals across cameras βœ… Production
face_detection/ Face detection & recognition pipeline (InsightFace) βœ… Production
embedding/ Feature extraction for visual similarity search βœ… Production
camera/ Camera capture utilities, RTSP/USB handlers βœ… Production
home-assistant-py/ Home Assistant integration (Python) βœ… Production
home-assistant-nodejs/ Home Assistant integration (Node.js) βœ… Production
fall_detection/ Fall detection for elderly care πŸ§ͺ Experimental
laptop_monitor/ Screen activity monitoring πŸ§ͺ Experimental
yolov2_parking/ Parking space detection (YOLOv2) πŸ“¦ Legacy
yolov3_parking/ Parking space detection (YOLOv3) πŸ“¦ Legacy
label-studio/ Label Studio annotation config πŸ“¦ Legacy
milvus/ Milvus vector database config πŸ“¦ Legacy
minio/ MinIO object storage config πŸ“¦ Legacy
flower/ Celery task monitor πŸ“¦ Legacy
monitor/ System monitoring utilities πŸ“¦ Legacy
build/ Build scripts for ARM/x86 platforms πŸ”§ Build
env/ Environment configuration πŸ”§ Build
unit_test/ Test suite πŸ”§ Test

Architecture

Camera Feed β†’ capture (camera/) β†’ detection (yolov7/) β†’ embedding (embedding/)
                                       ↓                        ↓
                                  face_detection/         yolov7_reid/
                                       ↓                        ↓
                                  Face Recognition      Person Re-ID
                                       β””β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                                            Notification
                                         (Home Assistant)

Evolution

These modules are being wrapped as skills for the SharpAI Aegis desktop app. The skill architecture provides a standard interface (SKILL.md) so each capability can be independently installed, configured, and updated.

Legacy Module New Skill
yolov7_person_detector/ skills/detection/yolo-detection-2026/
yolov7_reid/ skills/detection/person-recognition/ (planned)
face_detection/ + embedding/ skills/detection/face-recognition/ (planned)
home-assistant-py/ skills/integrations/homeassistant-bridge/ (planned)

See docs/skill-development.md for how to create new skills.