SYSTEMS AND METHODS FOR EXECUTING ATTENTION-BASED OBJECT SEARCHES ON IMAGES USING NEURAL NETWORK ARCHITECTURES

    公开(公告)号:US20220292129A1

    公开(公告)日:2022-09-15

    申请号:US17694517

    申请日:2022-03-14

    IPC分类号: G06F16/53 G06N3/08

    摘要: A systems including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform acts of: receiving a query image comprising at least one target object and at least one candidate image; generating a plurality of proposals based on the at least one candidate image; selecting a portion of the plurality of proposals to produce a reduced proposal set; generating a query embedding corresponding to the query image; generating candidate embeddings corresponding to the portion of the plurality of proposals included in the reduced proposal set; computing similarity scores for the portion of the plurality of proposals; and comparing the similarity scores to a threshold to determine if the at least one candidate image comprises the at least one target object. Other embodiments are disclosed herein.

    Systems and methods for executing attention-based object searches on images using neural network architectures

    公开(公告)号:US11815942B2

    公开(公告)日:2023-11-14

    申请号:US17694517

    申请日:2022-03-14

    IPC分类号: G06F16/53 G06N3/08

    CPC分类号: G06F16/53 G06N3/08

    摘要: A systems including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform acts of: receiving a query image comprising at least one target object and at least one candidate image; generating a plurality of proposals based on the at least one candidate image; selecting a portion of the plurality of proposals to produce a reduced proposal set; generating a query embedding corresponding to the query image; generating candidate embeddings corresponding to the portion of the plurality of proposals included in the reduced proposal set; computing similarity scores for the portion of the plurality of proposals; and comparing the similarity scores to a threshold to determine if the at least one candidate image comprises the at least one target object. Other embodiments are disclosed herein.