US 12,218,932 B2
Facial recognition tokenization
Margaret Inez Salter, Orlando, FL (US); Iustina-Miruna Vintila, Bucharest (RO); Arya Pourtabatabaie, Orlando, FL (US); Edison U. Ortiz, Orlando, FL (US); Sara Zafar Jafarzadeh, Kitchener (CA); Sayedmasoud Hashemi Amroabadi, Toronto (CA); and Christopher Côté Srinivasa, Toronto (CA)
Assigned to ROYAL BANK OF CANADA, Toronto (CA)
Filed by ROYAL BANK OF CANADA, Toronto (CA)
Filed on Jul. 21, 2021, as Appl. No. 17/382,255.
Claims priority of provisional application 63/110,214, filed on Nov. 5, 2020.
Claims priority of provisional application 63/054,630, filed on Jul. 21, 2020.
Prior Publication US 2022/0029987 A1, Jan. 27, 2022
Int. Cl. H04L 9/40 (2022.01); H04L 9/32 (2006.01)
CPC H04L 63/0861 (2013.01) [H04L 9/3218 (2013.01); H04L 9/3271 (2013.01); H04L 63/0435 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system for enhancing biometric template security, the system comprising:
a computer memory operating in conjunction with non-transitory computer readable data storage media housing at least a first data repository and a second data repository;
one or more computer processors configured to:
receive a data object representative of a full biometric feature set;
store a subset of the full biometric feature set or representations of a model trained from the full biometric feature set as a first partial feature or partial model portion set data object in the first data repository;
store a remaining subset of the full biometric feature set or representations of the model trained from the full biometric feature set in the second data repository, said remaining subset of said full biometric feature set or representations of said model trained from said full biometric feature set being accessible from said second data repository only via one or more zero-knowledge proof protocol interfaces; and
discard the data object representative of the full biometric feature set;
wherein the remaining subset selected for storage in the second data repository includes a plurality of feature or model representations exhibiting a largest variance in a training data set, wherein the plurality of feature representations exhibiting the largest variance are identified using one or more neural networks, each of said one or more neural networks having one or more controllable layers that are systematically deactivated to identify changes in classification accuracy, the systematic deactivation of the layers utilized to identify the features having the largest variance;
wherein both the first partial feature or partial model portion set data object and the remaining subset of the full biometric feature set or representations of the model are used during an authentication challenge to generate a matching response signal based on a data comparison against a new set of full features using (i) the first partial feature or partial model portion set data object and (ii) against the subset selected for storage in the second data repository interacted only through using the one or more zero-knowledge proof protocol interfaces.