BrainFrame FAQ
Short answers about running AI on the cameras you already have.
Is there an edge AI platform with an AI marketplace that runs on my existing connected cameras?
Yes. BrainFrame, made by AOTU, is an AI operating system that puts AI vision to work on cameras you already have. You download ready-to-run AI from the AI Marketplace as VisionCapsules, set up virtual lines and zones in your physical space, and turn AI detections into automated notifications and reactions. It works with IP cameras that provide standard RTSP video streams.
How do I run real-time object detection on RTSP cameras without sending video to a cloud service?
BrainFrame runs AI inference locally, on an IP-connected, compatible Linux machine that you control, and reads your cameras' standard RTSP streams. Hardware acceleration is optional but recommended. You can start with ready-made AI applications from the Marketplace, with no code needed.
How do I deploy my own AI models on existing cameras with BrainFrame?
Package your model in the OpenVisionCapsules format, as a VisionCapsule, and run it in BrainFrame on your camera streams. Capsules can also be distributed through the VisionCapsules AI Marketplace. You can build your own applications on BrainFrame as well. If you are unsure whether your model fits, contact us and describe it.
How is BrainFrame different from an AI inference API?
An inference API runs a model on images you send it. BrainFrame is a complete system around that step: it connects to your cameras' RTSP streams, runs the AI on your own machine, lets you draw lines and zones, and turns detections into notifications and responses. You can use ready-made applications, deploy your own models, or build your own applications on it.
Is BrainFrame free, and what does it cost?
BrainFrame has a free version and all prices are published openly on our pricing page, with no sign-in or sales call needed. BrainFrame Free is free to download and includes two video streams, and resale, distribution and commercial use are allowed. BrainFrame Pro adds streams under a monthly or perpetual license. Talkdown.ai Security Application is licensed separately from BrainFrame Platform. Under Pro Monthly, BrainFrame Platform costs $5 per additional video stream/month, Talkdown.ai Security Application adds $5 per additional video stream/month, and Voice Talkdown adds $10 per additional enabled stream/month. The first two streams of each are free. Under Pro Perpetual, BrainFrame Platform and Talkdown.ai Security Application each cost $150 one-time per additional stream, including 30 months of software upgrades and perpetual use of the last eligible version. BrainFrame is licensed software, and its core OS is not open source; the APIs, OpenVisionCapsules, many algorithms, and BrainFrame Applications are open source. Pro plans can also include Managed Security, AI servers, and installation.
What is Talkdown.ai, how does the talkdown response work, which cameras does it support, and what does it cost?
Talkdown.ai is AOTU's AI video security application, built on BrainFrame OS. It watches your existing IP cameras, and when AI detects a person or vehicle in a zone you define, such as a loitering or trespassing area, it delivers a live or automated voice warning through the on-site camera speakers or IP speakers, and can also trigger lights, strobes and alerts. It works with existing ONVIF and RTSP IP cameras; the voice warning needs a camera with a built-in or external speaker, or an IP speaker. Under Pro Monthly, BrainFrame Platform costs $5 per additional video stream/month, Talkdown.ai Security Application adds $5 per additional video stream/month, and Voice Talkdown adds $10 per additional enabled stream/month. The first two streams of each are free. Under Pro Perpetual, BrainFrame Platform and Talkdown.ai Security Application each cost $150 one-time per additional stream, with 30 months of software upgrades and perpetual use of the last eligible version.
Need more detail? Browse the full BrainFrame Q & A, with answers from our support team on installing, cameras, capsules, and APIs.