# US 12,223,722 B2 · Person identification systems and methods > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,223,722 B2 - Document ID: US12223722B2 - Status: Granted - Kind: B2 - Application number: 17/057,353 - Filed: 2018-05-25 - Published: 2021-06-24 - Granted: 2025-02-11 - Earliest claimed: 2018-05-25 - Assignees: Koninklijke Philips N.V. - Applicants: Koninklijke Philips N.V. - Inventors: Christine Menking Swisher; Purnima Rajan; Asif Rahman; Bryan Conroy - Web page: https://patentgrep.com/patent/US12223722B2 ## CPC - G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) - G06F 18/22: Matching criteria, e.g. proximity measures - G06V 10/26: Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Techniques disclosed herein relate to identifying individuals in digital images. In some embodiments, a digital image may be acquired (802) that captures an environment containing at least a first subject. A first portion of the digital image depicting the first subject may be segmented (806) into a plurality of superpixels. For each superpixel of the plurality of superpixels: a semantic label may be assigned (810) to the superpixel; features of the superpixel may be extracted (812); and a measure of similarity between the features extracted from the superpixel and features extracted from a reference superpixel identified in a reference digital image may be determined (814), wherein the reference superpixel has a reference semantic label that matches the semantic label assigned to the superpixel. Based on the measures of similarity associated with the plurality of superpixels, it may be determined (818) that the first subject is depicted in the reference image. ## Claims ### Claim 1 A computer-implemented method, comprising: acquiring, by one or more processors, a digital image that captures an environment containing at least a first subject; segmenting, by one or more of the processors, a first portion of the digital image depicting the first subject into a plurality of superpixels; for each superpixel of the plurality of superpixels: assigning, by one or more of the processors, a semantic label to the superpixel, extracting, by one or more of the processors, features of the superpixel, and determining, by one or more of the processors, a measure of similarity between the features extracted from the superpixel and features extracted from a reference superpixel identified in a reference digital image, wherein the reference superpixel has a reference semantic label that matches the semantic label assigned to the superpixel; and determining, by one or more of the processors, based on the measures of similarity associated with the plurality of superpixels, that the first subject is depicted in the reference image. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the digital image captures a scene containing a plurality of subjects that includes the first subject, and the method further comprises segmenting the digital image into a plurality of portions that each depicts a different subject of the plurality of subjects. ### Claim 3 (depends on 2) The computer-implemented method of claim 2, wherein the digital image is segmented into the plurality of portions by instance segmenting the plurality of subjects contained in the digital image into a mask. ### Claim 4 (depends on 3) The computer-implemented method of claim 3, wherein cuts of the mask are used to separate the mask into discrete sub-masks, wherein each sub-mask corresponds to an individual subject of the plurality of subjects. ### Claim 5 (depends on 2) The computer-implemented method of claim 2, wherein the digital image is segmented into the plurality of portions using pose estimation. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein at least one of the plurality of superpixels is assigned a semantic label that is associated with clothing worn over human torsos. ### Claim 7 (depends on 6) The computer-implemented method of claim 6, further comprising determining, by one or more of the processors, based on one or more of the measures of similarity associated with the plurality of superpixels, a likelihood that the first subject altered clothing worn over a torso of the first subject during a time interval between when the reference digital image was captured and when the digital image was captured. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, wherein at least one of the plurality of superpixels is assigned a semantic label that is associated with clothing worn over human legs. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein at least one of the plurality of superpixels is assigned a semantic label that is associated with human hair. ### Claim 10 (depends on 1) The computer-implemented method of claim 1, wherein at least one of the plurality of superpixels is assigned a semantic label that is associated with human faces. ### Claim 11 (depends on 6) The computer-implemented method of claim 6, further comprising determining, by one or more of the processors, based on the measures of similarity associated with the plurality of superpixels: a first likelihood that an appearance of the first subject was altered during a time interval between when the reference digital image was captured and when the digital image was captured, and a second likelihood that the first subject left the environment. ### Claim 12 (depends on 1) The computer-implemented method of claim 1, wherein the environment comprises a hospital waiting room, a gym, or an airport. ### Claim 13 A system configured to: acquire a digital image that captures an environment containing at least a first subject; segment a first portion of the digital image depicting the first subject into a plurality of superpixels; for each superpixel of the plurality of superpixels: assign a semantic label to the superpixel, extract features of the superpixel, and determine a measure of similarity between the features extracted from the superpixel and features extracted from a reference superpixel identified in a reference digital image, wherein the reference superpixel has a reference semantic label that matches the semantic label assigned to the superpixel; and determine, based on the measures of similarity associated with the plurality of superpixels, that the first subject is depicted in the reference image. ### Claim 14 (depends on 13) The system of claim 13, wherein the digital image captures a scene containing a plurality subjects that includes the first subject, and the system further comprises instructions to segment the digital image into a plurality of portions that each depicts a different subject of the plurality of subjects. ### Claim 15 At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to perform the following operations: acquiring a digital image that captures an environment containing at least a first subject; segmenting a first portion of the digital image depicting the first subject into a plurality of superpixels; for each superpixel of the plurality of superpixels: assigning a semantic label to the superpixel, extracting features of the superpixel, and determining a measure of similarity between the features extracted from the superpixel and features extracted from a reference superpixel identified in a reference digital image, wherein the reference superpixel has a reference semantic label that matches the semantic label assigned to the superpixel; and determining, based on the measures of similarity associated with the plurality of superpixels, that the first subject is depicted in the reference image.