Back to results
US 11,106,897 B2Granted
Human facial detection and recognition system
AssigneeBlue Line Security Solutions LLC
InventorMarcos Silva
Life of this patent
Estimating the term…
Abstract
Aspects of the present disclosure provide an image-based face detection and recognition system that processes and/or analyzes portions of an image using “image strips” and cascading classifiers to detect faces and/or various facial features, such an eye, nose, mouth, cheekbone, jaw line, etc.
Claims15
A computing device comprising:
- memory storing computer-readable instructions;
- and
- at least one processor to execute the computer-readable instructions to:
- receive video data from at least one IP camera, the video data comprising a sequence of images;
- detect by a detection component at least a portion of a face from the sequence of images;
- generate by the detection component a mapping of facial points corresponding to the at least the portion of the face from the sequence of images by generating a plurality of points on an image plane corresponding to the sequence of images;
- determine by the detection component that the sequence of images was taken at an angle and determine that there are three-dimensional pose variations in the sequence of images;
- cross-reference the mapping of facial points with each facial image of a set of pre-stored and previously recognized facial images to perform one of recognizing the face and not recognizing the face by a recognition component, the recognition component to generate a three-dimensional face model by mapping a two-dimensional representation of the facial points detected and mapped by the detection component to the three-dimensional facial model and generate three-dimensional feature points representing the three-dimensional face model, each feature point having a plurality of complex values, and group the feature points into a left half and a right half to result in two feature vectors for each face image;
- and
- automatically generate information associated with the face and transmit the information associated with the face.
7 more elements
A method comprising:
- receiving, by at least one processor, video data from at least one IP camera, the video data comprising a sequence of images;
- detecting, by a detection component, at least a portion of a face from the sequence of images;
- generating, by the detection component, a mapping of facial points corresponding to the at least the portion of the face from the sequence of images by generating a plurality of points on an image plane corresponding to the sequence of images;
- determining, by the detection component, that the sequence of images was taken at an angle and determining that there are three-dimensional pose variations in the sequence of images;
- cross-referencing the mapping of facial points with each facial image of a set of pre-stored and previously recognized facial images to perform one of recognizing the face and not recognizing the face by a recognition component, the recognition component generating a three-dimensional face model by mapping a two-dimensional representation of the facial points detected and mapped by the detection component to the three-dimensional facial model and generating three-dimensional feature points representing the three-dimensional face model, each feature point having a plurality of complex values, and grouping the feature points into a left half and a right half to result in two feature vectors for each face image;
- and
- automatically generating information associated with the face and transmitting the information associated with the face.
4 more elements
A non-transitory computer-readable medium having instructions stored thereon, the instructions when executed by at least one processor, cause the processor to perform operations comprising:
- receiving video data from at least one IP camera, the video data comprising a sequence of images;
- detecting, by a detection component, at least a portion of a face from the sequence of images;
- generating, by the detection component, a mapping of facial points corresponding to the at least the portion of the face from the sequence of images by generating a plurality of points on an image plane corresponding to the sequence of images;
- determining, by the detection component, that the sequence of images was taken at an angle and determining that there are three-dimensional pose variations in the sequence of images;
- cross-referencing the mapping of facial points with each facial image of a set of pre-stored and previously recognized facial images to perform one of recognizing the face and not recognizing the face by a recognition component, the recognition component generating a three-dimensional face model by mapping a two-dimensional representation of the facial points detected and mapped by the detection component to the three-dimensional facial model and generating three-dimensional feature points representing the three-dimensional face model, each feature point having a plurality of complex values, and grouping the feature points into a left half and a right half to result in two feature vectors for each face image;
- and
- automatically generating information associated with the face and transmitting the information associated with the face.
4 more elements
Dependencies are inferred from the claim text.