# US 2022/0284635 A1 · Method for making montage based on dialogue and apparatus using the same > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2022/0284635 A1 - Document ID: US20220284635A1 - Status: Published application - Kind: A1 - Application number: 17/513,629 - Filed: 2021-10-28 - Published: 2022-09-08 - Earliest claimed: 2021-03-02 - Assignees: Electronics and Telecommunications Research Institute - Applicants: Electronics and Telecommunications Research Institute - Inventors: Won-Joo PARK - Web page: https://patentgrep.com/patent/US20220284635A1 ## CPC - G06F 40/35: Discourse or dialogue representation - G06T 11/00: Two-dimensional [2D] image generation - G06F 40/279: Recognition of textual entities - G06N 3/045: Combinations of networks - G06N 3/0475: Generative networks - G06N 3/094: Adversarial learning - G06V 10/95: {structured as a network, e.g. client-server architectures} - G06V 40/169: {Holistic features and representations, i.e. based on the facial image taken as a whole} - G06V 40/171: {Local features and components; Facial parts (eye characteristics G06V40/18); Occluding parts, e.g. glasses; Geometrical relationships} - G06F 40/284: Lexical analysis, e.g. tokenisation or collocates - G06N 5/041: {Abduction} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Disclosed herein are a method for making a montage based on a dialogue and an apparatus using the same. The method includes extracting a feature for making a montage based on the dialogue between at least one inquirer and at least one respondent, generating a facial feature image based on the feature, and making a montage based on the facial feature image. ## Claims ### Claim 1 A method for making a montage based on a dialogue, comprising: extracting a feature for making a montage based on a dialogue between at least one inquirer and at least one respondent; generating a facial feature image based on the feature; and making a montage based on the facial feature image. ### Claim 2 (depends on 1) The method of claim 1, further comprising: determining whether there are missing features based on the dialogue; and generating an additional question for getting an additional feature when it is determined that there are missing features. ### Claim 3 (depends on 2) The method of claim 2, wherein the montage is made by combining multiple facial feature images generated based on multiple features. ### Claim 4 (depends on 1) The method of claim 1, further comprising: displaying the montage on a display shared between the at least one inquirer and the at least one respondent. ### Claim 5 (depends on 1) The method of claim 1, wherein extracting the feature comprises extracting a keyword corresponding to the feature by analyzing an answer provided by the at least one respondent, corresponding to a purport of a question asked by the at least one inquirer, and taking the keyword as metadata. ### Claim 6 (depends on 5) The method of claim 5, wherein generating the facial feature image comprises: retrieving the facial feature image from a facial feature image database in consideration of metadata corresponding to the feature; and generating a new facial feature image corresponding to the feature when a facial feature image tagged with the metadata is not present in the facial feature image database. ### Claim 7 (depends on 6) The method of claim 6, wherein generating the new facial feature image comprises generating the new facial feature image by applying a Generative Adversarial Network (GAN) to the facial feature image database. ### Claim 8 (depends on 6) The method of claim 6, wherein generating the new facial feature image comprises tagging the new facial feature image with the metadata and storing the new facial feature image tagged with the metadata in the facial feature image database. ### Claim 9 (depends on 6) The method of claim 6, further comprising: creating metadata by analyzing characteristics of each of multiple 2D or 3D facial feature images; and configuring the facial feature image database by tagging each of the multiple facial feature images with metadata corresponding thereto. ### Claim 10 (depends on 1) The method of claim 1, wherein the feature corresponds to characteristic information of each of facial components constituting the montage. ### Claim 11 (depends on 4) The method of claim 4, wherein displaying the montage comprises displaying at least one recommended montage as well as the montage. ### Claim 12 An apparatus for making a montage based on a dialogue, comprising: a processor for extracting a feature for making a montage based on a dialogue between at least one inquirer and at least one respondent, generating a facial feature image based on the feature, and making a montage based on the facial feature image; and a facial feature image database for managing the facial feature image. ### Claim 13 (depends on 12) The apparatus of claim 12, wherein the processor determines whether there are missing features based on the dialogue, and generates an additional question for getting an additional feature when it is determined that there are missing features. ### Claim 14 (depends on 12) The apparatus of claim 12, wherein the montage is made by combining multiple facial feature images generated based on multiple features. ### Claim 15 (depends on 12) The apparatus of claim 12, wherein the processor displays the montage on a display shared between the at least one inquirer and the at least one respondent. ### Claim 16 (depends on 12) The apparatus of claim 12, wherein the processor extracts a keyword corresponding to the feature by analyzing an answer provided by the at least one respondent, corresponding to a purport of a question asked by the at least one inquirer, and takes the keyword as metadata. ### Claim 17 (depends on 16) The apparatus of claim 16, wherein the processor retrieves the facial feature image from the facial feature image database in consideration of metadata corresponding to the feature, and when a facial feature image tagged with the metadata is not present in the facial feature image database, the processor generates a new facial feature image corresponding to the feature. ### Claim 18 (depends on 17) The apparatus of claim 17, wherein the processor generates the new facial feature image by applying a Generative Adversarial Network (GAN) to the facial feature image database. ### Claim 19 (depends on 17) The apparatus of claim 17, wherein the processor tags the new facial feature image with the metadata and stores the new facial feature image tagged with the metadata in the facial feature image database. ### Claim 20 (depends on 17) The apparatus of claim 17, wherein the facial feature image database is configured by tagging each of multiple 2D or 3D facial feature images with metadata created by analyzing characteristics of the facial feature image.