BrainFrame is an AI-powered video analytic platform that easily deploys AI models and algorithms to analyze camera footage and gather insights.
BrainFrame follows an iterative process for model training, combining pre-trained models, synthetic data, and customer-provided data to improve computer vision pipelines.
Deep dives
Brain Frame: Simplifying Video Analytics with AI
Brain Frame is an AI-powered video analytic platform that connects to existing camera hardware and extracts valuable information. It was created to address the challenge of changing customer requirements in the video analytics space. With Brain Frame, users can easily deploy AI models and algorithms to analyze camera footage and gather insights. It is designed to be user-friendly, catering to both technical algorithm developers and less technical system integrators who can work with the platform without writing code. Brain Frame offers a convenient solution for businesses to leverage computer vision and improve decision-making.
The Process of Model Training and Capsules in Brain Frame
Brain Frame follows an iterative process for model training, starting with a minimum viable model for basic recognition and then continuously refining it by collecting real-world data and labeling instances. The platform relies on a combination of pre-trained models, synthetic data, and customer-provided data to train and improve models. The use of capsules, which are standardized algorithms, allows for easy deployment and customization of computer vision pipelines. Capsules can be easily combined and configured to detect specific objects or features in video streams.
On-Premises Deployment and Cost Optimization
Brain Frame offers the flexibility of on-premises deployment, making efficient use of existing camera hardware. By utilizing local machines with GPUs, the platform minimizes the cost of model training. Training models is not a continuous process, so it is more cost-effective to have dedicated on-premises hardware rather than relying on cloud-based GPU instances. However, data labeling and storage are often performed in the cloud to take advantage of scalability and workforce availability.
Privacy Concerns and the Future of Computer Vision
The widespread adoption of cameras as sensors for computer vision raises concerns about privacy and surveillance. The future of computer vision will likely be influenced by the regulation of camera usage and public sentiment regarding constant monitoring. Brain Frame emphasizes maintaining privacy by focusing on metadata extraction rather than recording and storing video footage. Striking the right balance between the potential benefits of computer vision and privacy protection will shape the future landscape of this technology.
Imagine a world where you own some sort of building whether that’s a grocery store, a restaurant, a factory… and you want to know how many people reside in each section of the store, or maybe how long did the average person wait to be seated or how long did it take the average factory worker to complete their assembly task.
Currently today these systems are either not using AI and instead use a mix of sensors and buttons to track certain actions or they do use AI but in a way that’s highly specific to their use case and hard to easily modify for new use cases that come down the line.
This is where BrainFrame comes in. BrainFrame is a tool that connects to all your on-prem cameras and lets you easily leverage AI models and business logic. Alex Thiele is the CTO of Aotu the company that makes BrainFrame and he joins me today to talk about BrainFrame and the vision for a future where computer vision can be run by anyone.
This episode is hosted by David Cohen. David is a Software Engineering Lead at LinkedIn where he works on backend applications and APIs that power their enterprise data systems. In his free time, he is an AI enthusiast and enjoys talking about all things Software. You can contact him on LinkedIn or Twitter.