Video Analytics & Computer Vision

As the demand for IoT Video devices increases, the huge amount of data streams can overload the IoT environment. This is especially so with video analytics applications dependent on high-quality video data streamed over the network.

AnyConnect Video Analytics provides solutions for hybrid deployment of IoT networks and edge analytics that could help to distribute the load, reduce latency, comply with data protection, increase security, better cost management, trigger event-based alerts faster through predictive models, and more.

1. Device triggered by motion, sound, facial recognition, and more
2. Stream data
3. Video analytics (in the cloud)
4. Alerts & notifications (event-based)

Video analytics in the cloud

With AnyConnect’s solution, your devices can analyze images at the edge and in the cloud. Our computer vision & machine learning APIs can be deployed at the endpoint with our Vision Helper Library and our REST Services provides video analytics in the cloud.

1a. Device triggered by motion, sound, facial recognition, and more.
1b. Video analytics (at the edge)
2. Stream data
3. Video analytics (in the cloud)
4. Alerts & notifications (event-based)

Video analytics at the edge, in the cloud

Some of the features we provide

Computer vision

AnyConnect has developed extensive and extensible APIs and solutions for computer vision. Depending on your deployment, our Helper Libraries can run at the edge (in cameras and mobile devices), in the cloud, and in the fog.

As they query from the same source, they share the common logic across, bringing the best of edge computing and cloud computing.

This makes our computer vision techniques scalable across general-purpose, low-power system-on-chip to dedicated, high-power computing platforms.

computer vision
Multiple perspectives

Single Scene / Multiple Perspectives

AnyConnect IoT Video platform powers cameras at multiple proximal locations, providing multiple perspectives of single scenes. We are able to combine different perspectives of single scenes for better computer vision.

Multiple Modalities Inputs

AnyConnect Computer Vision techniques combine multiple modalities inputs including audio and video for better computer vision.

Start building your IoT Video devices with us.