Ingesting streaming signals
First Claim
Patent Images
1. A method comprising:
- concurrently during ingestion of a raw streaming signal from a streaming signal source;
sampling a single frame contained in the raw streaming signal;
separating the raw streaming signal into a plurality of segments; and
storing the plurality of segments at a storage device;
computing a preliminarily severity of a possible real-world event of a real-world event type from a hint with respect to the more complete content of the raw streaming signal included in the sampled single frame;
triggering a deeper inspection of the raw streaming signal based on the preliminary severity;
performing the deeper inspection of the raw streaming signal, including;
requesting the deeper inspection;
playing a segment, from among the plurality of segments, from the storage device;
inspecting the content of the segment during play of the segment;
computing an actual severity of the real-world event of the real-world event type based on the preliminary severity and the inspected content; and
updating a normalized Time, Location, Context (TLC) signal corresponding to the raw streaming signal to include the actual severity.
1 Assignment
0 Petitions
Accused Products
Abstract
The present invention extends to methods, systems, and computer program products for ingesting streaming signals. Signal ingestion modules sample a frame from a raw streaming signal. A preliminary severity or possible event type is computed from the sampled frame. A deeper inspection request is triggered of the raw streaming signal or of another raw streaming signal. Segments of content from the raw streaming signal or the other raw streaming signal are inspected. An actual severity or actual event type is computed from the inspected content. The actual severity or actual event type can be included in a normalized signal.
11 Citations
20 Claims
-
1. A method comprising:
-
concurrently during ingestion of a raw streaming signal from a streaming signal source; sampling a single frame contained in the raw streaming signal; separating the raw streaming signal into a plurality of segments; and storing the plurality of segments at a storage device; computing a preliminarily severity of a possible real-world event of a real-world event type from a hint with respect to the more complete content of the raw streaming signal included in the sampled single frame; triggering a deeper inspection of the raw streaming signal based on the preliminary severity; performing the deeper inspection of the raw streaming signal, including; requesting the deeper inspection; playing a segment, from among the plurality of segments, from the storage device; inspecting the content of the segment during play of the segment; computing an actual severity of the real-world event of the real-world event type based on the preliminary severity and the inspected content; and updating a normalized Time, Location, Context (TLC) signal corresponding to the raw streaming signal to include the actual severity. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. A method comprising:
-
sampling a single frame contained in a raw streaming signal from a streaming signal source; preliminarily classifying the raw streaming signal as indicative of a real-world event type through reference to a content classifier and from a hint with respect to the more complete content of the raw streaming signal included in the sampled single frame; triggering a deeper inspection of the raw streaming signal based on the preliminary classification; performing the deeper inspection of the raw streaming signal, including; requesting the deeper inspection; playing a portion of the raw streaming signal from the streaming signal source; inspecting the content of the portion during play of the portion; and computing a probability that the raw streaming signal is actually indicative of the real-world event type; and updating a context dimension of a normalized Time, Location, Context (TLC) signal corresponding to the raw streaming signal to include the probability of the real-world event type. - View Dependent Claims (9, 10, 11, 12, 13)
-
-
14. A method comprising:
-
sampling a single frame contained in a first raw streaming signal from a streaming signal source; preliminarily classifying the first raw streaming signal as indicative of a real-world event type through reference to a content classifier and from a hint with respect to the more complete content of the first raw streaming signal included in the sampled single frame; triggering a deeper inspection of a second raw signal based on the preliminary classification; performing the deeper inspection of the second raw signal, including; requesting the deeper inspection; playing a segment from among a plurality of segments of the second raw streaming signal; inspecting content of the segment during replay of the segment; and computing a probability that the first raw streaming signal is actually indicative of a real-world event of the real-world event type based on the inspected content; and updating a context dimension of a normalized Time, Location, Context (TLC) signal corresponding to the first raw streaming signal to include the probability of the real-world event type. - View Dependent Claims (15, 16, 17, 18, 19, 20)
-
Specification