Back to results
US 12,762,892 B2Granted
Apparatus for identifying a face and method thereof
AssigneesHyundai Motor Company, Kia Corporation
InventorsKeon Ho Lee, Moon Sub Jin, Kyu Sang Choi, Hyeon Seok Jung, Kyung Hwan Cho
Life of this patent
Estimating the term…
Abstract
The present disclosure relates to a face identification apparatus and a method thereof. An apparatus includes: a processor configured to detect face information from image data, to extract face features from the detected face information based on a face feature extraction model weight-lightened through a weight-lightening technique, and to compare the face features with a previously stored face database to identify a face according to similarity; and a storage configured to store data and algorithms driven by the processor, and to have the face database.
Claims20
A face identification apparatus comprising:
- an image sensor configured to capture an image of a face of a user and generate image data corresponding to the captured image;
- a processor configured to:
- detect face information from the image data;
- extract face features from the detected face information based on a face feature extraction model weight-lightened through a weight-lightening technique;
- compare the face features with face data of a previously stored face database, wherein the face data comprise face features respectively associated with a plurality of user identifiers (IDs);
- identify, based on a face similarity result associated with a comparison of the face features and the face data, the face;
- determine a total score for each user identifier (ID) based on a sum of scores associated with face features for the respective user ID, wherein the scores are assigned to the respective user ID based on the face similarity result between the extracted face features and the face features associated with the respective user ID stored in the face database;
- and
- output, as an identification result, information of a user ID having a highest total score among the user IDs by:
- based on at least two user IDs having a same total score, outputting, as the identification result, a user ID having a smaller number of scores among the at least two user IDs;
- and
- a storage configured to store data and algorithms executed by the processor, and to have the face database.
9 more elements
A face identification method comprising:
- capturing, by an image sensor, an image of a face of a user and generating image data corresponding to the captured image;
- detecting, by a processor, face information from the image data;
- extracting, by the processor, face features from the detected face information based on a face feature extraction model weight-lightened through a weight-lightening technique;
- comparing, by the processor, the extracted face features with face data of a previously stored face database, wherein the face data comprise face features respectively associated with a plurality of user identifiers (IDs);
- identifying, based on a face similarity result associated with the comparing, the face;
- determining a total score for each user identifier (ID) based on a sum of scores associated with face features for the respective user ID, wherein the scores are assigned to the respective user ID based on the face similarity result between the extracted face features and the face features associated with the respective user ID stored in the face database;
- and
- outputting, as an identification result, information of a user ID having a highest total score among the user IDs by:
- based on at least two user IDs having a same total score, outputting, as the identification result, a user ID having a smaller number of scores among the at least two user IDs.
6 more elements
A face identification method performed by an apparatus, the face identification method comprising:
- capturing, by an image sensor, an image of a face of a user and generating image data corresponding to the captured image;
- pre-processing, by a processor, the image data by performing an alignment associated with a direction of the face;
- detecting face information from the pre-processed image data;
- generating a weight-lightened face feature extraction model by:
- performing a network pruning on a trained face feature extraction model, and
- retraining the network-pruned face feature extraction model;
- extracting, by the processor, face features from the detected face information by:
- inputting, parameters of the detected face information, as an input to the network-pruned face feature extraction model;
- comparing, by the processor, the extracted face features with face data of a previously stored face database;
- and
- outputting, based on a face similarity result associated with the comparing, a face identification result corresponding to a user identifier (ID) associated with face features stored in the face database.
8 more elements
Dependencies are inferred from the claim text.