Interactive video surveillance as an edge service using unsupervised feature queries

An intermediate node for a distributed sensor network, comprising:

• a communication port configured to communicate with an automated communication network;
• a memory;
• and
• at least one automated processor, configured to:
• control the communication port to receive a communication representing data from a sensor node;
• determine characteristics of the data using at least one of machine learning and statistical inference;
• storing the feature data in the memory;
• receive a query through the communication port;
• determine a correspondence of the query to the characteristics;
• and
• releasing data from the sensor node selectively in dependence on the correspondence of the query to the characteristics.

A system comprising:

• a plurality of cameras, each camera of the plurality of cameras having a distinct geolocation;
• at least one computing device in electronic communication with each of the plurality of cameras, the at least one computing device being configured to determine whether an object of interest is present in frames captured by at least one camera of the plurality of cameras, the at least one computing device being configured to:
• generate object data relating to at least one object included in a frame captured by each of the plurality of cameras;
• receive a query describing an object of interest;
• determine if the object of interest is included in any of the frames captured by each of the plurality of cameras based on the generated object data;
• identify at least one matching frame captured by at least one camera of the plurality of cameras that includes the object of interest;
• and
• provide match data relating to at least one of:
• the at least one matching frame including the object of interest defined in the query, or
• the at least one camera of the plurality of cameras that captured the at least one matching frame including the object of interest defined in the query.