US 2019/0294881 A1Published application

Behavior recognition

AssigneeViisights Solutions Ltd.

InventorsSimon Polak, Menashe Rothschild, Asaf Birenzvieg

Life of this patent

Estimating the term…

Loading patent…

Abstract

A computer implemented method of identifying a behavior of objects detected in a video stream, comprising using one or more processors adapted for receiving a video stream comprising a plurality of frames captured by imaging sensor(s), applying a first set of trained classification functions for visually analyzing each of at least a subset of the frames to create a semantic attribute set for each object detected in each frame comprising an object probability score indicative of the object class probability, attribute probability score(s) indicative of probability of attribute(s) associated with the object and estimated location of the object, applying a second set of trained classification functions for semantically analyzing the attribute sets created for each object tracked in the subset to calculate a behavior probability score indicative of action(s) estimated to be conducted by each object and outputting an indication when the calculated behavior probability score complies with predefined rule(s).

Claims13

Dependent claims

A computer implemented method of identifying a behavior of objects detected in a video stream, comprising:

using at least one processor adapted for:
receiving a video stream comprising a plurality of frames captured by at least one imaging sensor;
applying a first set of trained classification functions for visually analyzing each of at least a subset of the plurality of frames to create a respective one of a plurality of semantic attribute sets for at least one object detected in a respective frame, the respective attribute set comprising an object probability score indicative of a probability of a class of the at least one object, an attribute probability score indicative of a probability of at least one of a plurality of attributes associated with the at least one object and an estimated location of the at least one object in the respective frame;
applying a second set of trained classification functions for semantically analyzing the plurality of attribute sets created for the at least one object tracked in the subset of frames to calculate a behavior type probability score indicative of at least one of a plurality of actions estimated to be conducted by the at least one object;
and
outputting an indication in case the calculated behavior type probability score complies with at least one predefined rule.

3 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/116,968Filed 30 Aug 2018
Priority: 22 Mar 2018
Published: 26 Sep 2019
Term ends: Not estimatedThe source records don’t support an estimate.

Classification24 codes

G06F 16/71
Of video data: indexing; Data structures therefor; Storage structures
G06F 18/214
Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation: generating training…
G06F 18/2415
Relating to the classification model, e.g. parametric or non-parametric approaches: based on parametric or probabilistic models, e.g. based on…
G06F 18/24317
Relating to the number of classes: piecewise classification, i.e. whereby each classification requires several discriminant rules