Before you begin, ensure your cameras are added and streaming properly. Right-click the main screen in Blue Iris, select "Add new camera," and input your camera's IP address, login credentials, and stream parameters.
The image is analyzed locally via computer vision models (such as YOLOv5 or YOLOv8). If CodeProject.AI matches the target with a high enough confidence score (e.g., person: 82% ), it passes a status back to Blue Iris. Only then is the alert officially verified and sent to your mobile device or smart home ecosystem. If nothing is found, the alert is cancelled. Step-by-Step Guide to Setting Up Verified Alerts
In Blue Iris → AI tab:
This means you can now run our AI models directly through CodeProject.AI on your Blue Iris NVR with full confidence in compatibility and performance. Say goodbye to cloud latency and hello to local, private, and fast object detection. codeproject blue iris verified
Ensure that local traffic on port 5000 is not being blocked.
Finally got CodeProject.AI and Blue Iris "Verified" – 100% Reliable Alerts!
Given the lack of specific context, here are a few possible interpretations: Before you begin, ensure your cameras are added
This deep integration is so "verified" that recent versions of Blue Iris no longer function with the old DeepStack method, solidifying CodeProject.AI as the future of the platform. According to discussions from the CodeProject community, "The last Blue Iris version that was not tightly integrated with CPAI in this way was 5.6.8.4". As Blue Iris evolves, its integration with CodeProject.AI becomes more intrinsic, making the "verified" setup the only officially supported path forward.
Download the latest CodeProject.AI Server installer for Windows. Run the installer with default options, which typically include modules like "Object Detection (YOLOv5 .NET)" and "Face Processing".
This is where the two programs connect. The integration is built directly into Blue Iris, making verification easy. If CodeProject
CPU usage spikes to 100%; inference time is > 500ms. Fix: In CodeProject.AI Server dashboard ( http://localhost:32168 ), check System Info . If your NVIDIA GPU is not listed, install the correct CUDA toolkit (v12.x). Restart the AI server.
By pairing , a leading Windows-based Network Video Recorder (NVR), with CodeProject.AI Server , a free, local, open-source AI platform, you can implement a "verified" alert architecture . This configuration ensures your system records continuously or logs motion locally, but only interrupts your day when a specific target—such as a person, vehicle, or animal—is successfully identified and verified by local computer vision. The Core Concept: How "Verified" Alerts Work