AN AI FABRIC SEARCH METHOD FOR SEARCHING FOR AND RECOMMENDING SIMILAR FABRICS BY USING A VIDEO RECORDING VARIOUS STATES OF CLOTHING FABRICS, A RECORDING MEDIUM STORING A PROGRAM FOR EXECUTING THE SAME, AND A COMPUTER PROGRAM STORED IN THE RECORDING MEDIUM FOR EXECUTING THE SAME

摘要

The present invention provides a property keyframe for a basic fabric image that can check the properties of a fabric from a video using a deep learning-based property extraction engine when receiving a video of continuous motion for checking various properties of a garment fabric; Extracts attribute keyframes for stretch, thickness, and wrinkled fabric images, and extracts attribute images that capture the degree to which the fabric is gradually stretched from a wrinkled state, and a feature image that captures the amount of jolly when the fabric is shaken left and right. Then, using a deep learning-based primary search engine, first search for fabrics similar to the basic fabric photo, rank them according to the degree of similarity, and then pull and stretch the fabric, the thickness of the fabric, and the image of the fabric being wrinkled. By correcting the rankings of similar fabrics corresponding to the primary search results by applying the video of the fabric to the extent to which it is slowly unfolded from a crumpled state, and the extent to which the fabric is jolly when shaken left and right to the deep learning-based secondary correction engine. , an artificial intelligence fabric search method that searches for and recommends similar fabrics using videos of various states of clothing fabrics that can search and recommend similar fabrics in consideration of various characteristics as well as basic shapes of fabrics. It relates to a recording medium in which a program is stored and a computer program stored in the recording medium to execute the same.