easyaiot  by soaring-xiongkulu

AIoT platform for intelligent vision and control

Created 6 months ago
270 stars

Top 95.4% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

EasyAIoT is an open-source, cloud-edge-device integrated IoT platform designed to democratize AI by enabling truly zero-barrier AI access for everyone. It focuses on massive camera deployments, merging real-time streaming with AI analytics for intelligent vision, making sophisticated AI capabilities accessible across diverse IoT scenarios.

How It Works

The platform utilizes a unique tri-lingual mixed programming architecture (Java, Python, C++) to optimize stability, AI development, and performance. It integrates AI and IoT across cloud, edge, and device layers, featuring multi-protocol camera access (GB28181, ONVIF), vision large model integration (QwenVL3), and millisecond-level real-time AI analysis. A key innovation is zero-shot labeling technology using large models for automated data labeling and model self-optimization.

Quick Start & Requirements

Automated installation scripts are provided for Linux, macOS, and Windows, alongside Docker containerized deployment options. Supports x86 and ARM (RK3588) architectures, with localized OS compatibility (Kylin, UOS).

  • Demo URL: http://36.111.47.113:8888/ (admin/admin123)
  • Documentation: Link to "Platform Deployment Documentation" mentioned.
  • Repositories: GitHub (https://github.com/soaring-xiongkulu/easyaiot), Gitee (https://gitee.com/soaring-xiongkulu/easyaiot).

Highlighted Details

  • AI: Supports GB28181/ONVIF cameras, QwenVL3 vision LLM, real-time analysis (detection, behavior), OCR, and speech recognition.
  • IoT: Comprehensive device management, MQTT/TCP/HTTP protocol support, rule engine, and multi-channel notifications.
  • Architecture: Modular (7 projects) for independent edge deployment.
  • Cross-platform: Linux (server/edge), macOS, Windows (incl. ARM).

Maintenance & Community

Contributors are noted for documentation and deployment scripts. Community access is facilitated through technical exchange groups, accessible via WeChat and an official account/knowledge planet.

Licensing & Compatibility

MIT License: 100% free use, modification, distribution for individuals/enterprises. No commercial restrictions.

Limitations & Caveats

Caveat: Explicitly a 'learning project,' not for commercial activities. Users bear full legal responsibility for compliance and usage risks.

Health Check
Last Commit

1 month ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
38 stars in the last 30 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), Soumith Chintala Soumith Chintala(Coauthor of PyTorch), and
1 more.

jetson-inference by dusty-nv

0.1%
9k
Vision DNN library for NVIDIA Jetson devices
Created 9 years ago
Updated 4 months ago
Feedback? Help us improve.