# US 12,118,769 B1 · Machine learning architecture for peer-based image scoring > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,118,769 B1 - Document ID: US12118769B1 - Status: Granted - Kind: B1 - Application number: 18/752,622 - Filed: 2024-06-24 - Published: 2024-10-17 - Granted: 2024-10-15 - Earliest claimed: 2017-07-26 - Assignees: Vizit Labs, Inc. - Applicants: Vizit Labs, Inc. - Inventors: Elham Saraee; Jehan Hamedi; Zachary Halloran - Web page: https://patentgrep.com/patent/US12118769B1 ## CPC - G06V 10/761: {Proximity, similarity or dissimilarity measures} - G06F 16/438: Presentation of query results - G06N 3/045: Combinations of networks - G06N 20/00: Machine learning - G06V 10/40: Extraction of image or video features - G06V 10/82: using neural networks - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A method includes displaying a first user interface including a target image and a task selection element, in response to a selection of the task selection element, displaying a second user interface including a target audience selection element, in response to a selection of a target audience using the target audience selection element, displaying a third user interface including a benchmark selection element, in response to a selection of a benchmark using the benchmark selection element, executing a target audience machine learning model using as input the target image to generate a performance score, calculating a benchmark score for the target image based on the performance score of the target image and performance scores for the images of the benchmark, and displaying a fourth user interface including the target image and the benchmark score. ## Claims ### Claim 1 A non-transitory, computer-readable medium including instructions which, when executed by one or more processors, cause the one or more processors to: display, via an interactive graphic user interface (GUI), a first user interface including a target image and a task selection element; in response to a selection of the task selection element, display, via the interactive GUI, a second user interface including a target audience selection element; in response to a selection of a target audience using the target audience selection element, display, via the interactive GUI, a third user interface including a benchmark selection element; in response to a selection of a benchmark using the benchmark selection element: execute a target audience machine learning model using as input the target image to generate a performance score indicating a performance of the target image for the target audience; calculate a benchmark score for the target image based on the performance score of the target image and performance scores for the images of the benchmark, the benchmark score indicating a performance of the target image for the target audience relative to the images of the benchmark; and display, via the interactive GUI, a fourth user interface including the target image and the benchmark score. ### Claim 2 (depends on 1) The non-transitory, computer-readable medium of claim 1, wherein the benchmark is generated using the target audience machine learning model. ### Claim 3 (depends on 2) The non-transitory, computer-readable medium of claim 2, wherein the instructions cause the one or more processors to generate the benchmark by executing the target audience machine learning model using as input the images of the benchmark to generate the performance scores for the images of the benchmark. ### Claim 4 (depends on 1) The non-transitory, computer-readable medium of claim 1, wherein the selection of the benchmark includes a selection of a first benchmark and a second benchmark, and wherein the fourth user interface includes a first benchmark score for the first benchmark and a second benchmark score for the second benchmark. ### Claim 5 (depends on 1) The non-transitory, computer-readable medium of claim 1, wherein the instructions cause the one or more processors to display, via the interactive GUI, the third user interface in response to a selection of a benchmark score generation task. ### Claim 6 (depends on 1) The non-transitory, computer-readable medium of claim 1, wherein the first user interface includes a plurality of target images including the target image, and wherein the fourth user interface includes the plurality of target images and corresponding benchmark scores. ### Claim 7 (depends on 6) The non-transitory, computer-readable medium of claim 6, wherein the instructions cause the one or more processors to, in response to the selection of the target audience, display, via the interactive GUI a fifth user interface including the plurality of target images ranked according to performance scores indicating a performance of the plurality of target images for the target audience. ### Claim 8 A non-transitory, computer-readable medium including instructions which, when executed by one or more processors, cause the one or more processors to: display, via an interactive graphic user interface (GUI), a first user interface including a target image; in response to a selection of the target image, display, via the interactive GUI, a second user interface including a target audience selection element; in response to a selection of a target audience using the target audience selection element, display, via the interactive GUI, a third user interface including a benchmark selection element; and in response to a selection of a benchmark using the benchmark selection element: execute a target audience machine learning model using as input the target image to generate a performance score indicating a performance of the target image for the target audience; and display, via the interactive GUI, a fourth user interface including the target image and benchmark images corresponding to the benchmark, the target image and the benchmark images ranked according to performance scores for the target audience including the generated performance score of the target image. ### Claim 9 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the benchmark is generated using the target audience machine learning model. ### Claim 10 (depends on 9) The non-transitory, computer-readable medium of claim 9, wherein the instructions cause the one or more processors to generate the benchmark by executing the target audience machine learning model using as input the images of the benchmark to generate performance scores for the images of the benchmark. ### Claim 11 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the first user interface includes a task selection element, and wherein the instructions cause the one or more processors to display, via the interactive GUI, the third user interface in response to a selection of a benchmark comparison task. ### Claim 12 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the first user interface includes a plurality of target images including the target image, and wherein the fourth user interface includes the plurality of target images and the benchmark images ranked according to performance for the target audience. ### Claim 13 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the fourth user interface includes a benchmark score for the target image, the benchmark score indicating performance of the target image for the target audience relative to the benchmark images. ### Claim 14 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the fourth user interface includes benchmark scores for the benchmark images. ### Claim 15 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the fourth user interface includes a first depiction of the target image ranked with the benchmark images and a second depiction of the target image separate from the benchmark images. ### Claim 16 A non-transitory, computer-readable medium including instructions which, when executed by one or more processors, cause the one or more processors to: display, via an interactive graphic user interface (GUI), a first user interface including a plurality of images and a task selection element; in response to a selection of a ranking task using the task selection element: execute a target audience machine learning model using as input the plurality of images to generate performance scores for the plurality of images indicating a performance of the plurality of images for the target audience; display, via the interactive GUI, a second user interface including the plurality of images ranked according to the performance scores; and store the performance scores of the plurality of images as a benchmark. ### Claim 17 (depends on 16) The non-transitory, computer-readable medium of claim 16, wherein the instructions cause the one or more processors to store the performance scores of the plurality of images as the benchmark in response to selection of a benchmark creation task. ### Claim 18 (depends on 17) The non-transitory, computer-readable medium of claim 17, wherein the benchmark creation task is selected using the task selection element. ### Claim 19 (depends on 17) The non-transitory, computer-readable medium of claim 17, wherein the instructions cause the one or more processors to display, via the interactive GUI, the first user interface in response to selection of the benchmark creation task. ### Claim 20 (depends on 16) The non-transitory, computer-readable medium of claim 16, wherein the first user interface includes an indication of the target audience.