Skip to content

Releases: EfficientTools/graffiti-detection-ai-model

v1.5.0

Choose a tag to compare

@github-actions github-actions released this 18 Jul 16:33

Full Changelog: v1.4.0...v1.5.0

v1.4.0

Choose a tag to compare

@github-actions github-actions released this 18 Jul 05:12

What's Changed

  • Bump actions/setup-python from 6.2.0 to 6.3.0 in the workflow-actions group by @dependabot[bot] in #1

New Contributors

Full Changelog: v1.3.0...v1.4.0

v1.3.0

Choose a tag to compare

@github-actions github-actions released this 18 Jul 00:16

Full Changelog: v0.1.3...v1.3.0

v1.2.0 Graffiti Detection AI Model Release (May 9th 2026

Choose a tag to compare

@pH-7 pH-7 released this 09 May 02:23

v1.2.0 — Graffiti Detection AI Model

What's New

This release brings a more robust end-to-end pipeline, improved alert handling, and better production readiness across the board.

Features

  • Multi-camera surveillance — monitor multiple camera streams simultaneously with multi_camera_surveillance.py
  • Alert channels — out-of-the-box support for email, SMS, webhook, Discord, and Slack notifications on detection events
  • FastAPI service — REST endpoints for single image, annotated, batch detection and live stats (/detect, /detect/annotated, /detect/batch, /stats)
  • Incident logging — SQLite/JSON persistence with daily reporting via incident_logger.py
  • Real-time dashboard — live stats overlay with real_time_dashboard.py
  • Data augmentation pipelinealbumentations-backed augmentation for training robustness
  • Street scenario configs — pre-built YAML configs for urban surveillance deployments

Improvements

  • Extended test coverage: alerts, augmentation, dataset, metrics, visualization, integration, and street scenarios
  • Docker / Docker Compose / Kubernetes deployment configs (see DEPLOYMENT.md)
  • Python 3.8–3.12 compatibility verified

Installation

pip install graffiti-detection-ai-model==1.2.0

Or upgrade from a previous version:

pip install --upgrade graffiti-detection-ai-model

Quick Start

from graffiti_detection import GraffitiDetector

detector = GraffitiDetector("models/best.pt")
detections = detector.predict("image.jpg")

☕️ Found this useful? Buy me a coffee to support more AI innovations!


Full Changelog: v1.1.0...v1.2.0

Graffiti Detection AI Model v1.1.0 (March 28th 2026 Release)

Choose a tag to compare

@pH-7 pH-7 released this 28 Mar 04:09

🚀 v1.1.0 Release Notes

Highlights

  • Added BlueSky badge for improved author social discoverability
  • Updated X badge label and all badge tooltips for cleaner, more accessible social links
  • New SVG logo for a modern project identity
  • FastAPI REST endpoint for graffiti detection
  • Real-time multi-camera surveillance system
  • Optimized model configs for better real-time performance
  • PyPI release pipeline for direct pip installation
  • Deployment docs, Docker/Compose setup, and expanded tests
  • Improved test compatibility across the data pipeline

Other Changes

  • Updated dependencies for API and alert system integration
  • Added LICENSE.md

Full Changelog: v1.0.0...v1.1.0

v1.0.0 - Initial Release (Jan 11th 2026)

Choose a tag to compare

@pH-7 pH-7 released this 14 Jan 10:53