IMAGE PROCESSING DEVICE AND IMAGE PROCESSING METHOD

    公开(公告)号:US20240303977A1

    公开(公告)日:2024-09-12

    申请号:US18443381

    申请日:2024-02-16

    发明人: Yasuhisa IKUSHIMA

    摘要: A processor: executes classification of classifying a plurality of validation images into a plurality of classes with a machine learning model trained with a plurality of training images; obtains a degree of separation between the plurality of classes by the classification of the plurality of validation images and evaluates accuracy of the classification of the plurality of validation images based on the obtained degree of separation between the plurality of classes; and evaluates whether re-training of the machine learning model is necessary based on an evaluation result of the accuracy of classification of the plurality of validation images, extracts an validation image whose classification result has a relatively high possibility to be erroneous from among the plurality of validation images to automatically re-train the machine learning model if it is evaluated that the re-training of the machine learning model is necessary.

    GENERATING AND USING BEHAVIORAL POLICY GRAPHS THAT ASSIGN BEHAVIORS TO OBJECTS FOR DIGITAL IMAGE EDITING

    公开(公告)号:US20240257421A1

    公开(公告)日:2024-08-01

    申请号:US18162401

    申请日:2023-01-31

    申请人: Adobe Inc.

    发明人: Kevin Gary Smith

    摘要: The present disclosure relates to systems, methods, and non-transitory computer-readable media that generate and implement behavioral policy graphs for digital image editing. For instance, in some embodiments, the disclosed systems generate, for a client device, a behavioral policy graph that assigns behaviors to object classes based on object relationships. The disclosed systems receive, from the client device, a digital image portraying a plurality of objects. Further, the disclosed systems determine behaviors of the plurality of objects utilizing the behavioral policy graph by determining, for each object of the plurality of objects, a behavior based on a relationship of the object with an additional object of the plurality of objects in accordance with the behavioral policy graph. The disclosed systems modify the digital image by modifying one or more objects based on the behaviors of the plurality of objects.

    INSPECTION SUPPORT DEVICE, INSPECTION SUPPORT METHOD, AND PROGRAM

    公开(公告)号:US20240219312A1

    公开(公告)日:2024-07-04

    申请号:US18606782

    申请日:2024-03-15

    发明人: Shuhei HORITA

    摘要: Provided are an inspection support device, an inspection support method, and a program capable of reducing complexity of creating a damage diagram. The inspection support device includes a processor that supports creation of a damage diagram of a structure, in which the processor acquires image data of information including a structural drawing of a target structure on a medium and damage information related to damage added by a user on the medium, recognizes the damage information by image recognition from the acquired image data, acquires drawing data corresponding to the structural drawing, aligns the structural drawing of the image data with the drawing data, and draws the damage information as a damage graphic at a corresponding position of the drawing data to create the damage diagram.

    ARTIFICIAL INTELLIGENCE DEVICE FOR ATTENTION OVER DETECTION BASED OBJECT SELECTION AND CONTROL METHOD THEREOF

    公开(公告)号:US20240346814A1

    公开(公告)日:2024-10-17

    申请号:US18635931

    申请日:2024-04-15

    摘要: A method for controlling an artificial intelligence (AI) device can include obtaining an input query, an input image, bounding boxes for objects detected in the input image, object labels corresponding to the bounding boxes, and at least one topic label for a word in the input query, generating at least one word embedding for the at least one topic label, and generating a plurality of word embeddings for the object labels corresponding to the bounding boxes. The method can further include generating output attention maps corresponding to scaled dot product attention matrices based on the at least one word embedding for the at least one topic label from the input query and each of the plurality of word embeddings for the object labels, and combining the output attention maps to generate a final attention map corresponding to the at least one topic label from the input query.