利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

19 条记录 (耗时 0.014 秒)

    Multi-modal differential search with real-time focus adaptation
    1.
    发明授权
    Multi-modal differential search with real-time focus adaptation 有权

    公开(公告)号：US11604822B2

    公开(公告)日：2023-03-14

    申请号：US16426369

    申请日：2019-05-30

    申请人： Adobe Inc.

    发明人： Pranav Vineet Aggarwal ,  Zhe Lin ,  Baldo Antonio Faieta ,  Saeid Motiian

    IPC分类号： G06F7/00 ,  G06F16/583 ,  G06N3/084 ,  G06N20/00 ,  G06F16/538 ,  G06F16/532 ,  G06F16/33 ,  G06F3/04855

    摘要： Multi-modal differential search with real-time focus adaptation techniques are described that overcome the challenges of conventional techniques in a variety of ways. In one example, a model is trained to support a visually guided machine-learning embedding space that supports visual intuition as to “what” is represented by text. The visually guided language embedding space supported by the model, once trained, may then be used to support visual intuition as part of a variety of functionality. In one such example, the visually guided language embedding space as implemented by the model may be leveraged as part of a multi-modal differential search to support search of digital images and other digital content with real-time focus adaptation which overcomes the challenges of conventional techniques.

    UNIFIED FRAMEWORK FOR MULTI-MODAL SIMILARITY SEARCH
    4.
    发明申请
    UNIFIED FRAMEWORK FOR MULTI-MODAL SIMILARITY SEARCH 有权

    公开(公告)号：US20220391450A1

    公开(公告)日：2022-12-08

    申请号：US17887694

    申请日：2022-08-15

    申请人： Adobe Inc.

    发明人： Pranav Vineet Aggarwal ,  Ali Aminian ,  Ajinkya Gorakhnath Kale ,  Aashish Kumar Misraa

    IPC分类号： G06F16/903 ,  G06N20/00 ,  G06F16/908 ,  G06K9/62

    摘要： Technology is disclosed herein for enhanced similarity search. In an implementation, a search environment includes one or more computing hardware, software, and/or firmware components in support of enhanced similarity search. The one or more components identify a modality for a similarity search with respect to a query object. The components generate an embedding for the query object based on the modality and based on connections between the query object and neighboring nodes in a graph. The embedding for the query object provides the basis for the search for similar objects

    Multi-Modal Differential Search with Real-Time Focus Adaptation
    5.
    发明申请
    Multi-Modal Differential Search with Real-Time Focus Adaptation 审中-公开

    公开(公告)号：US20200380027A1

    公开(公告)日：2020-12-03

    申请号：US16426369

    申请日：2019-05-30

    申请人： Adobe Inc.

    发明人： Pranav Vineet Aggarwal ,  Zhe Lin ,  Baldo Antonio Faieta ,  Saeid Motiian

    IPC分类号： G06F16/583 ,  G06N3/08 ,  G06N20/00 ,  G06F16/33 ,  G06F16/538 ,  G06F16/532

    摘要： Multi-modal differential search with real-time focus adaptation techniques are described that overcome the challenges of conventional techniques in a variety of ways. In one example, a model is trained to support a visually guided machine-learning embedding space that supports visual intuition as to “what” is represented by text. The visually guided language embedding space supported by the model, once trained, may then be used to support visual intuition as part of a variety of functionality. In one such example, the visually guided language embedding space as implemented by the model may be leveraged as part of a multi-modal differential search to support search of digital images and other digital content with real-time focus adaptation which overcomes the challenges of conventional techniques.

    Text-to-Visual Machine Learning Embedding Techniques
    7.
    发明申请
    Text-to-Visual Machine Learning Embedding Techniques 有权

    公开(公告)号：US20210365727A1

    公开(公告)日：2021-11-25

    申请号：US17398317

    申请日：2021-08-10

    申请人： Adobe Inc.

    发明人： Pranav Vineet Aggarwal ,  Zhe Lin ,  Baldo Antonio Faieta ,  Saeid Motiian

    IPC分类号： G06K9/62 ,  G06F16/535 ,  G06N20/00 ,  G06K9/72

    摘要： Text-to-visual machine learning embedding techniques are described that overcome the challenges of conventional techniques in a variety of ways. These techniques include use of query-based training data which may expand availability and types of training data usable to train a model. Generation of negative digital image samples is also described that may increase accuracy in training the model using machine learning. A loss function is also described that also supports increased accuracy and computational efficiency by losses separately, e.g., between positive or negative sample embeddings a text embedding.

    Text-to-visual machine learning embedding techniques
    8.
    发明授权
    Text-to-visual machine learning embedding techniques 有权

    公开(公告)号：US11144784B2

    公开(公告)日：2021-10-12

    申请号：US16426264

    申请日：2019-05-30

    申请人： Adobe Inc.

    发明人： Pranav Vineet Aggarwal ,  Zhe Lin ,  Baldo Antonio Faieta ,  Saeid Motiian

    IPC分类号： G06K9/62 ,  G06K9/72 ,  G06F16/535 ,  G06N20/00 ,  G06F3/0482

    摘要： Text-to-visual machine learning embedding techniques are described that overcome the challenges of conventional techniques in a variety of ways. These techniques include use of query-based training data which may expand availability and types of training data usable to train a model. Generation of negative digital image samples is also described that may increase accuracy in training the model using machine learning. A loss function is also described that also supports increased accuracy and computational efficiency by losses separately, e.g., between positive or negative sample embeddings a text embedding.

    Text-to-Visual Machine Learning Embedding Techniques
    9.
    发明申请
    Text-to-Visual Machine Learning Embedding Techniques 审中-公开

    公开(公告)号：US20200380298A1

    公开(公告)日：2020-12-03

    申请号：US16426264

    申请日：2019-05-30

    申请人： Adobe Inc.

    发明人： Pranav Vineet Aggarwal ,  Zhe Lin ,  Baldo Antonio Faieta ,  Saeid Motiian

    IPC分类号： G06K9/62 ,  G06K9/72 ,  G06F16/535 ,  G06N20/00

    摘要： Text-to-visual machine learning embedding techniques are described that overcome the challenges of conventional techniques in a variety of ways. These techniques include use of query-based training data which may expand availability and types of training data usable to train a model. Generation of negative digital image samples is also described that may increase accuracy in training the model using machine learning. A loss function is also described that also supports increased accuracy and computational efficiency by losses separately, e.g., between positive or negative sample embeddings a text embedding.