# US 8,542,888 B2 · Detecting facial similarity based on human perception of facial similarity > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,542,888 B2 - Document ID: US8542888B2 - Status: Granted - Kind: B2 - Application number: 13/549,959 - Filed: 2012-07-16 - Published: 2012-11-08 - Granted: 2013-09-24 - Earliest claimed: 2006-05-10 - Assignees: AOL Inc. - Applicants: Sharon M. Perlmutter; Keren O. Perlmutter; Joshua Alspector; Alex Holub; Mark Everingham; Pietro Perona; Andrew Zisserman - Inventors: Sharon M. Perlmutter; Keren O. Perlmutter; Joshua Alspector; Alex Holub; Mark Everingham; Pietro Perona; Andrew Zisserman - Web page: https://patentgrep.com/patent/US8542888B2 ## CPC - G06V 40/172: {Classification, e.g. identification} - G06F 18/2163: {Partitioning the feature space} - G06F 18/2178: {based on feedback of a supervisor} - G06F 18/40: Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor - G06V 10/763: {Non-hierarchical techniques, e.g. based on statistics of modelling distributions} - G06V 10/7784: {based on feedback from supervisors} - G06V 10/945: {User interactive design; Environments; Toolboxes} - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Similar faces may be determined within images based on human perception of facial similarity. The user may provide an image including a query face to which the user wishes to find faces that are similar. Similar faces may be determined based on similarity information. Similarity information may be generated from information related to a human perception of facial similarity. Images that include faces determined to be similar, based on the similarity information, may be provided to the user as search result images. The user then may provide feedback to indicate the user's perception of similarity between the query face and the search result images. ## Claims ### Claim 1 A method for determining facial similarity among images in a social networking service, the method comprising: determining, using at least one processor, inequalities between a plurality of training faces associated with one or more images based on received indications of similarity between a plurality of image groupings; determining, using the at least one processor, a mapping function between an objective feature space and a psychological feature space that indicates human perception of facial similarity based on the inequalities; receiving a query associated with an image that includes a query face; determining, using the at least one processor, one or more search result images from among a plurality of images associated with users of a social networking service based on the mapping function; and providing the search result images to a user of the social networking service. ### Claim 2 (depends on 1) The method of claim 1 wherein: the query comprises receiving an image for analysis from the user; and the one or more search result images comprise images associated with at least one online photo album and the user. ### Claim 3 (depends on 2) The method of claim 2 wherein: the one or more search result images further comprise images associated with at least one online photo album and other users of the social networking service. ### Claim 4 (depends on 1) The method of claim 1, further comprising: receiving a selection made by the user of characteristics associated with a person in the image that includes a query face; and removing search result images from the determined one or more search result images based on the received selection. ### Claim 5 (depends on 1) The method of claim 1, further comprising: detecting one or more classes associated with the query face; identifying characteristics of the query face based on at least the determined one or more classes; and reducing a field of search associated with the received query based on at least on the identified characteristics. ### Claim 6 (depends on 1) The method of claim 1, further comprising: identifying a plurality of search image results from among the provided search image results that was accessed by the user; determining one or more common characteristic of the plurality of accessed search image results; updating the search results based at least on the determined one or more common characteristics; and providing the updated search results images to the user. ### Claim 7 (depends on 1) The method of claim 1, further comprising: receiving an identifier associated with the query face; and determining one or more search result images based on the received identifier. ### Claim 8 A system for determining facial similarity among images within a social networking service, comprising: a memory device storing a plurality of instructions; and at least one processor that executes the plurality of instructions to cause the at least one processor to: determine inequalities between a plurality of training faces associated with one or more images based on received indications of similarity between a plurality of image groupings; determine a mapping function between an objective feature space and a psychological feature space that indicates human perception of facial similarity based on the inequalities; receive a query associated with an image that includes a query face; determine one or more search result images from among a plurality of images associated with users of a social networking service based on the mapping function; and provide the search result images to a user of the social networking service. ### Claim 9 (depends on 8) The system of claim 8, wherein: the query comprises receiving an image for analysis from the user; and the one or more search result images comprise images associated with at least one online photo album and the user. ### Claim 10 (depends on 9) The system of claim 9, wherein: the one or more search result images further comprise images associated at least one online photo album and other users of the social networking service. ### Claim 11 (depends on 8) The system of claim 8, wherein the at least one processor further executes the plurality of instructions to: receive a selection made by the user of characteristics associated with a person in the image that includes a query face; and remove search result images from the determined one or more search result images based on the received selection. ### Claim 12 (depends on 8) the system of claim 8, wherein the at least one processor further executes the plurality of instructions to: detect one or more classes associated with the query face; identify characteristics of the query face based on at least the determined one or more classes; and reduce a field of search associated with the received query based on at least on the identified characteristics. ### Claim 13 (depends on 8) The system of claim 8, wherein the at least one processor further executes the plurality of instructions to: identify a plurality of search image results from among the provided search image results that was accessed by the user; determine one or more common characteristic of the plurality of accessed search image results; update the search results based at least on the determined one or more common characteristics; and provide the updated search results images to the user. ### Claim 14 (depends on 8) The system of claim 8, wherein the at least one processor further executes the plurality of instructions to: receive an identifier associated with the query face; and determine one or more search result images based on the received identifier. ### Claim 15 A non-transitory computer-useable medium storing a computer program, the computer program including a set of instructions that are executable by at least one processor to cause the at least processor to: determine inequalities between a plurality of training faces associated with one or more images based on received indications of similarity between a plurality of image groupings; determine a mapping function between an objective feature space and a psychological feature space that indicates human perception of facial similarity based on the inequalities; receive a query associated with an image that includes a query face; determine one or more search result images from among a plurality of images associated with users of a social networking service based on the mapping function; and provide the search result images to a user of the social networking service. ### Claim 16 (depends on 15) The medium of claim 15, wherein: the query comprises receiving an image for analysis from the user; and the one or more search result images comprise images associated with at least one online photo album and the user. ### Claim 17 (depends on 16) The medium of claim 16, wherein: the one or more search result images further comprises images associated at least one online photo album and other users of the social networking service. ### Claim 18 (depends on 15) The medium of claim 15, wherein the set of instructions further cause the at least one processor to: receive a selection made by the user of characteristics associated with a person in the image that includes a query face; and remove search result images from the determined one or more search result images based on the received selection. ### Claim 19 (depends on 15) The medium of claim 15, wherein the set of instructions further cause the at least one processor to: detect one or more classes associated with the query face; identify characteristics of the query face based on at least the determined one or more classes; and reduce a field of search associated with the received query based on at least on the identified characteristics. ### Claim 20 (depends on 15) The medium of claim 15, wherein the set of instructions further cause the at least one processor to: identify a plurality of search image results from among the provided search image results that was accessed by the user; determine one or more common characteristic of the plurality of accessed search image results; update the search results based at least on the determined one or more common characteristics; and provide the updated search results images to the user. ### Claim 21 (depends on 15) The medium of claim 15, wherein the set of instructions further cause the at least one processor to: receive an identifier associated with the query face; and determine one or more search result images based on the received identifier.