Overview

What is VisionAIre Stream?

An on-premise vision AI to empower your camera. This product is supersede its predecessor, Visionaire Cluster, with only single-line instruction and can run on a casual gaming laptop.

Why Visionaire Docker Stream?

Flexible Deployment: Our platform offers superior ease of deployment, both on-premises on your own servers and in cloud environments with ease.
Efficient Installation: Simple designed installation process, allowing for quick and easy installation.
Advanced Video Processing: the ability to process camera acquisitions in real-time for streaming data via RTSP and ONVIF protocols, as well as pre-recorded video, thus offering transmission in your use.
Extensive Real Time Analytics Collection: Our platform is equipped with a wide range of analytics options, from facial recognition to license plate detection, providing a comprehensive solution to your every problem.
Camera Agnostic Capability: Our system is designed for compatibility with a wide range of camera types, eliminating dependency on specific brands.
Flexible Custom Solutions: Visionaire offers plug-and-integrate capabilities for custom-developed analytics solutions, tailored to your specific needs.
Easy Integration: Our platform seamlessly integrates with various applications via REST API and WebSocket, ensuring smooth data management and usage.

How does Visionaire Docker Stream work?

VisionAIre Docker Stream utilizes a combination of other services:

Postgres - For database,
Docker Snapshot - For enrichment model service,
Visionaire Docker Stream - For video stream processing and main model service,
Vanilla Dashboard (Optional) - Built-in dashboard for visualization.

How our System Integrated

CCTV Stream Consumption: The system starts with CCTV cameras capturing video. These video streams are then consumed by the VisionAIre Stream using RTSP (Real Time Streaming Protocol) or RTMP (Real Time Messaging Protocol) addresses. This allows the VisionAIre Stream to access the live video feeds.
Image Processing and AI Detection: VisionAIre Stream takes these video feeds and performs a series of tasks such as encoding and decoding images. VisionAIre Stream processes them using AI models to detect various objects or events. This could involve recognizing faces, detecting intruders, or any other security-related events. Once processed, the data from these events are stored in a database.
Database Integration: The processed data is stored in a database for further use. This database can be queried to retrieve historical data for analysis or reporting.
Snapshot Storage: VisionAIre Snapshot component is involved in storing 'enrichment data'. This refers to additional data that enriches from the stream data.
Secondary AI Inference: The VisionAIre Stream sends cropped images from the initial detection model to the VisionAIre Snapshot for a second round of AI inference. For instance, if the first model detects a person, the cropped image of the person is sent to the VisionAIre Snapshot where it may be processed by a more specialized AI model to detect attributes of the person, such as age, gender, clothing, etc. The results from this second inference are then sent back to the VisionAIre Stream.
Dashboard Alerting and Real-time Display: The data from the VisionAIre Stream is also sent to a dashboard. This dashboard serves as a visualization tool for real-time data and alerting. If an event is detected that requires immediate attention, such as unauthorized access, the dashboard will provide an alert. Additionally, it displays real-time data, which could include live feeds, detection events, or statistics.
Licensing:
In the stream license, we have 2 different options that are differentiated by time limit.
1. Perpetual License. So when users buy this kind of licenses, then the analytics will be active until our product becomes legacy. But for this type of license, we only provide 1 major update free of charge if the user doesn’t buy our support package.
2. Time Based License. This license is used for most of our projects, we usually sell it yearly based on license. But we can adjust the time duration for the license from day, month to a year. If users decide to buy with this license, they will get a free update as long as the subscription is still active.

PreviousAbout Nodeflux NextRequirements

Last updated 1 year ago

Was this helpful?

What is VisionAIre Stream?

An on-premise vision AI to empower your camera. This product is supersede its predecessor, Visionaire Cluster, with only single-line instruction and can run on a casual gaming laptop.

Why Visionaire Docker Stream?

Flexible Deployment: Our platform offers superior ease of deployment, both on-premises on your own servers and in cloud environments with ease.

Efficient Installation: Simple designed installation process, allowing for quick and easy installation.

Advanced Video Processing: the ability to process camera acquisitions in real-time for streaming data via RTSP and ONVIF protocols, as well as pre-recorded video, thus offering transmission in your use.

Extensive Real Time Analytics Collection: Our platform is equipped with a wide range of analytics options, from facial recognition to license plate detection, providing a comprehensive solution to your every problem.

Camera Agnostic Capability: Our system is designed for compatibility with a wide range of camera types, eliminating dependency on specific brands.

Flexible Custom Solutions: Visionaire offers plug-and-integrate capabilities for custom-developed analytics solutions, tailored to your specific needs.

Easy Integration: Our platform seamlessly integrates with various applications via REST API and WebSocket, ensuring smooth data management and usage.

How does Visionaire Docker Stream work?

VisionAIre Docker Stream utilizes a combination of other services:

Postgres - For database,

Docker Snapshot - For enrichment model service,

Visionaire Docker Stream - For video stream processing and main model service,

Vanilla Dashboard (Optional) - Built-in dashboard for visualization.

How our System Integrated

CCTV Stream Consumption: The system starts with CCTV cameras capturing video. These video streams are then consumed by the VisionAIre Stream using RTSP (Real Time Streaming Protocol) or RTMP (Real Time Messaging Protocol) addresses. This allows the VisionAIre Stream to access the live video feeds.

Image Processing and AI Detection: VisionAIre Stream takes these video feeds and performs a series of tasks such as encoding and decoding images. VisionAIre Stream processes them using AI models to detect various objects or events. This could involve recognizing faces, detecting intruders, or any other security-related events. Once processed, the data from these events are stored in a database.

Database Integration: The processed data is stored in a database for further use. This database can be queried to retrieve historical data for analysis or reporting.

Snapshot Storage: VisionAIre Snapshot component is involved in storing 'enrichment data'. This refers to additional data that enriches from the stream data.

Secondary AI Inference: The VisionAIre Stream sends cropped images from the initial detection model to the VisionAIre Snapshot for a second round of AI inference. For instance, if the first model detects a person, the cropped image of the person is sent to the VisionAIre Snapshot where it may be processed by a more specialized AI model to detect attributes of the person, such as age, gender, clothing, etc. The results from this second inference are then sent back to the VisionAIre Stream.

Dashboard Alerting and Real-time Display: The data from the VisionAIre Stream is also sent to a dashboard. This dashboard serves as a visualization tool for real-time data and alerting. If an event is detected that requires immediate attention, such as unauthorized access, the dashboard will provide an alert. Additionally, it displays real-time data, which could include live feeds, detection events, or statistics.

Licensing:

In the stream license, we have 2 different options that are differentiated by time limit.

Perpetual License. So when users buy this kind of licenses, then the analytics will be active until our product becomes legacy. But for this type of license, we only provide 1 major update free of charge if the user doesn’t buy our support package.
Time Based License. This license is used for most of our projects, we usually sell it yearly based on license. But we can adjust the time duration for the license from day, month to a year. If users decide to buy with this license, they will get a free update as long as the subscription is still active.