Back to results
US 11,443,551 B2Granted
Facial recognitions based on contextual information
AssigneeHewlett-Packard Development Company, L.P.
InventorsHarold Merkel, Qian Lin, Guoxing Yang
Life of this patent
Estimating the term…
Abstract
In example implementations, a method is provided. The method may be executed by a processor. The method includes receiving an image. The image is analyzed to obtain facial features. Contextual information is obtained and a vector including a facial feature class of the facial features and contextual feature classes of the contextual information is generated. A facial recognition is then performed based on the vector.
Claims14
A method, comprising:
- receiving, by a processor, an image;
- analyzing, by the processor, the image to obtain facial features;
- obtaining, by the processor, contextual information;
- generating, by the processor, a vector comprising a facial feature class of the facial features and contextual feature classes of the contextual information;
- accessing, by the processor, dynamic contextual information stored in dynamic contextual information databases to update dynamic contextual information of the contextual feature classes that has changed for a respective vector of each image of a plurality of images in a database in response to a request to perform a facial recognition process, wherein the dynamic contextual information comprises calendar information and access event data, wherein the access event data includes an access card swipe or a log in of a remote device to an internet protocol address at a location;
- comparing, by the processor, the vector with the respective vector of each image of the plurality of images in the database that are updated;
- and
- detecting, by the processor, a matching image from the plurality of images in the database based on a highest score calculated from each matching feature between the vector and the respective vector of the matching image.
5 more elements
A non-transitory computer readable storage medium encoded with instructions executable by a processor, the non-transitory computer-readable storage medium comprising:
- instructions to receive a facial image;
- instructions to analyze the facial image to calculate facial features;
- instructions to access a database to obtain static contextual information and dynamic contextual information;
- instructions to generate a vector comprising a facial feature class of the facial features and contextual feature classes of the static contextual information and the dynamic contextual information;
- instructions to access the dynamic contextual information stored in dynamic contextual information databases to update the dynamic contextual information of the contextual feature classes that has changed for a respective vector of each image of a plurality of facial images in a gallery database in response to a request to perform a facial recognition process, wherein the dynamic contextual information comprises calendar information and access event data, wherein the access event data includes an access card swipe or a log in of a remote device to an internet protocol address at a location;
- instructions to compare the vector with the respective vector of each facial image of the plurality of facial images in the gallery database that are updated;
- and
- instructions to detect a matching facial image from the plurality of facial images in the gallery database based on a highest score calculated from each matching feature between the vector and the respective vector of the matching facial image.
5 more elements
An apparatus, comprising:
- a camera to capture a facial image;
- a first database to store a plurality of facial images;
- a second database to contextual information;
- and
- a processor communicatively coupled to the camera, the first database, and the second database, wherein the processor receives the facial image, calculates facial features from the facial image, obtains contextual information from the second database, generates a vector comprising a facial feature class of the facial features and contextual feature classes of the contextual information, accesses dynamic contextual information stored in dynamic contextual information databases to update dynamic contextual information of the contextual feature classes that has changed for a respective vector of each image of a plurality of facial images in a gallery database in response to a request to perform a facial recognition process, wherein the dynamic contextual information comprises calendar information and access event data, wherein the access event data includes an access card swipe or a log in of a remote device to an internet protocol address at a location, compares the vector with the respective vector of each facial image of the plurality of facial images in the gallery database that are updated, and detects a matching facial image from the plurality of facial images in the gallery database based on a highest score calculated from each matching feature between the vector and the respective vector of the matching facial image.
2 more elements
Dependencies are inferred from the claim text.