Abstract
A system and method are provided for enhancing the operational intelligence of network-connected cameras through the establishment of camera-specific semantic context. A digital repository is associated with an individual monitoring device to store context metadata representing physical and logical relationships in the device's field of view. The context metadata can be defined via user input or automatically inferred using analytical models that identify historical activity patterns or process environmental feedback. When a monitoring event occurs, the associated context metadata is retrieved and integrated into an evaluation framework, such as an augmented prompt for a generative artificial intelligence model, to analyze the video stream. This allows the system to distinguish between relevant events and benign activities, thereby reducing false notifications and improving video captioning accuracy. The digital repository remains fully visible and editable by a user to maintain data accuracy and user control. Keywords: semantic awareness, camera context, notification filtering, smart home security, context repository, generative model prompt.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Och, Josh; Axley, Andrew; and , "Context-Aware Semantic Method And System For Network-Connected Camera Notification Filtering", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/11603