Methods circuits devices systems and associated computer executable code for video feed processing
Disclosed are methods, circuits, devices, systems and associated executable code for multi factor image feature registration and tracking, wherein utilized factors include both static and dynamic parameters within a video feed. Assessed factors may originate from a heterogeneous set of sensors including both video and audio sensors. Acoustically acquired scene information may supplement optically acquired information.
1. A video feed processing system comprising:
interface circuits to receive one or more two dimensional (“2D”) video feeds;
processing circuitry executing instructions to:
track and characterize moving elements present within the scene video feed;
identify and characterize static elements or objects present within the scene video feed;
define a three dimensional (“3D”) coordinate set for the scene within boundaries of the scene video feed, wherein defining a three dimensional (“3D”) coordinate set includes identifying one or more substantially straight edges, indicative of the X Y or Z axes of the scene and mapping the positions of the identified and characterized static elements or objects in relation to the identified straight edges;
derive coordinates and orientation of the video feed source within the scene video feed coordinate frame;
update a scene element inventory with characterization information relating to identified static elements, wherein updating is performed responsive to either newly derived characterization information or to refinement of existing characterization information;
detect from within the tracked and characterized moving elements one or more humans in the scene video feed;
characterize and log an event associated with a detected human; and
further comprising an echo-ranging assembly in proximity with the video feed source and including one or more acoustic emitters and one or more acoustic detectors.
2. The system according to claim 1 , wherein static element characterization information for a static element includes one or more of: (a) estimated dimensions of an element; (b) an element category; (c) estimated coordinates of one or more points on the static element; (d) a relationship with other elements in the scene; and (e) scene presence parameters.
3. The system according to claim 1 , wherein characterization of a given static element is at least partially derived from detected interactions between the given static element and a tracked moving element.
4. The system according to claim 1 , wherein scene boundary coordinates are at least partially derived from detected interactions between a surface and a tracked moving element.
5. The system according to claim 1 , wherein characterization of a moving element detected as a human includes one or more of: (a) a human category; (b) a human's recognized identity; (c) a human's group affiliation or membership; and (d) a human's relationship to other humans identified in the scene.
6. The system according to claim 1 , wherein the processing circuitry is further adapted to:
detect from within the tracked and characterized moving elements one or more non-human entities in the scene video feed, and wherein non-human entities are selected from a group consisting of: animals, mechanically moving devices, static objects or plants affected by their environment, shadows, and displayed or projected images; and
characterize and log an event associated with a detected non-human entity.
7. The system according to claim 1 , wherein said echo-ranging assembly includes a combination of at least: one of said emitters and two of said detectors, or one of said detectors and two of said emitters, utilized by said processing circuitry to perform stereo acoustic triangulation and derive directional information pertaining to the source of an acoustic flashback detected by at least one of said acoustic detectors.
8. The system according to claim 1 , wherein said acoustic emitter and or said acoustic detector are multi-directional and said processing circuitry correlates a distance associated with a round-trip time of each of a set of detected echoes with each of a set of visually identified static elements or each of a set of tracked moving objects.
9. The system according to claim 1 , wherein said processing circuitry generates a references echo map when no moving objects are detected with the scene.
10. The system according to claim 1 , wherein said processing circuitry generates a references echo map when one or more moving objects are detected within the scene.
11. The system according to claim 1 , wherein deriving video feed source coordinates and orientation includes comparing an appearance of a given identified static element within the video feed against reference data corresponding to the identified object.
12. A method of video feed processing, said method system comprising:
electronically receiving one or more two dimensional (“2D”) video feeds;
algorithmically tracking and characterizing moving elements present within the scene video feed;
algorithmically identifying and characterizing static elements or objects present within the scene video feed;
defining a three dimensional (“3D”) coordinate set for the scene within boundaries of the scene video feed, wherein defining a three dimensional (“3D”) coordinate set includes identifying one or more substantially straight edges, indicative of the X Y or Z axes of the scene and mapping the positions of the identified and characterized static elements or objects in relation to the identified straight edges, wherein mapping the positions of the identified and characterized static elements or objects includes automatically echo-ranging one or more of the identified and characterized static elements or objects;
deriving coordinates and orientation of the video feed source within the scene video feed coordinate frame;
updating a scene element inventory with characterization information relating to identified static elements, wherein updating is performed responsive to either newly derived characterization information or to refinement of existing characterization information;
detecting from within the tracked and characterized moving elements one or more humans in the scene video feed; and
characterizing and logging an event associated with a detected human.
13. The method according to claim 12 , wherein static element characterization information for a static element includes one or more of: (a) estimated dimensions of an element; (b) an element category; (c) estimated coordinates of one or more points on the static element; (d) a relationship with other elements in the scene; and (e) scene presence parameters.
14. The method according to claim 12 , wherein characterization of a given static element is at least partially derived from detected interactions between the given static element and a tracked moving element.
15. The method according to claim 12 , wherein scene boundary coordinates are at least partially derived from detected interactions between a surface and a tracked moving element.
16. The method according to claim 12 , wherein characterization of a moving element detected as a human includes one or more of: (a) a human category; (b) a human's recognized identity; (c) a human's group affiliation or membership; and (d) a human's relationship to other humans identified in the scene.
17. The method according to claim 12 , further including detecting from within the tracked and characterized moving elements one or more non-human entities in the scene video feed, and wherein non-human entities are selected from a group consisting of: animals, mechanically moving devices, static objects or plants affected by their environment, shadows, and displayed or projected images; and
characterize and log an event associated with a detected non-human entity.