Plumerai Custom AI Notifications¶
With Plumerai Custom AI Notifications, users can specify open-ended, text-based scenarios for which they would like to receive notifications. For example, users can specify "notify me when a person is watering the plants" or "notify me when a cat is on the fence".
Plumerai Custom AI Notifications is built using a Vision Language Model ('VLM') consisting of a video embedder and text embedder. It runs accurately and efficiently because it leverages all the power of Plumerai's TinyML solutions, including Plumerai Object Detection, Plumerai Familiar Face Identification, and Plumerai Advanced Motion Detection.
Product features:
- Leverages Plumerai's TinyML and VLM combo, resulting in high accuracy and significant cloud compute cost reductions.
- Can run efficiently on the edge, fully in the cloud, or in a hybrid set-up.
Custom AI Notifications architecture¶
Plumerai Custom AI Notifications is built upon the same building blocks as Plumerai Video Search, and also consists of four main components, each described in more detail below and illustrated in the following diagram.
The first two components, 'VLM Video Collection' and 'VLM Video Embedder' run continuously on video data provided by the camera. These components are designed to be efficient and fast and can run on an edge device.
The last two components, 'VLM Text Embedder' and 'Video and Text Matcher' run once per defined notification alert, typically on-demand based on user-input. These can run in the cloud or on a user's mobile phone or laptop for example.
The four main components are described on the Plumerai Video Search product page in more detail.
The API is not yet public, but can be made available upon request.
Deployment scenarios¶
Similar to Plumerai Video Search, there are various deployment scenarios possible for Plumerai Custom AI Notifications. We refer to the Plumerai Video Search product page for an idea what kind of deployments are possible.