利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

414 条记录 (耗时 0.033 秒)

    Generating object masks of object parts utlizing deep learning
    2.
    发明授权
    Generating object masks of object parts utlizing deep learning 有权

    公开(公告)号：US11900611B2

    公开(公告)日：2024-02-13

    申请号：US18147278

    申请日：2022-12-28

    申请人： Adobe Inc.

    发明人： Yinan Zhao ,  Brian Price ,  Scott Cohen

    IPC分类号： G06T7/11 ,  G06T7/168 ,  G06T9/00 ,  G06T5/00 ,  G06T7/181

    CPC分类号： G06T7/11 ,  G06T5/004 ,  G06T7/168 ,  G06T7/181 ,  G06T9/002 ,  G06T9/008

    摘要： The present disclosure relates to a class-agnostic object segmentation system that automatically detects, segments, and selects objects within digital images irrespective of object semantic classifications. For example, the object segmentation system utilizes a class-agnostic object segmentation neural network to segment each pixel in a digital image into an object mask. Further, in response to detecting a selection request of a target object, the object segmentation system utilizes a corresponding object mask to automatically select the target object within the digital image. In some implementations, the object segmentation system utilizes a class-agnostic object segmentation neural network to detect and automatically select a partial object in the digital image in response to a target object selection request.

    SEMANTICALLY-AWARE IMAGE EXTRAPOLATION
    3.
    发明公开
    SEMANTICALLY-AWARE IMAGE EXTRAPOLATION 审中-公开

    公开(公告)号：US20230169632A1

    公开(公告)日：2023-06-01

    申请号：US17521503

    申请日：2021-11-08

    申请人： Adobe Inc.

    发明人： Kuldeep Kulkarni ,  Soumya Dash ,  Hrituraj Singh ,  Bholeshwar Khurana ,  Aniruddha Mahapatra ,  Abhishek Bhatia

    IPC分类号： G06T5/50 ,  G06T7/181

    CPC分类号： G06T5/50 ,  G06T7/181

    摘要： Certain aspects and features of this disclosure relate to semantically-aware image extrapolation. In one example, an input image is segmented to produce an input segmentation map of object instances in the input image. An object generation network is used to generate an extrapolated semantic label map for an extrapolated image. The extrapolated semantic label map includes instances in the original image and instances that will appear in an outpainted region of the extrapolated image. A panoptic label map is derived from coordinates of output instances in the extrapolated image and used to identify partial instances and boundaries. Instance-aware context normalization is used to apply one or more characteristics from the input image to the outpainted region to maintain semantic continuity. The extrapolated image includes the original image and the outpainted region and can be rendered or stored for future use.

    Edge detection system and methods
    10.
    发明授权
    Edge detection system and methods 有权

    公开(公告)号：US09697434B2

    公开(公告)日：2017-07-04

    申请号：US14566308

    申请日：2014-12-10

    申请人： OmniVision Technologies, Inc.

    发明人： Robbert Emery

    IPC分类号： G06K9/46 ,  G06T7/12 ,  G06T7/181

    CPC分类号： G06K9/4609 ,  G06T7/12 ,  G06T7/181 ,  G06T2207/10024

    摘要： An edge detection method includes reading at least a portion of wide-band pixel signals generated by wide-band pixels of an image sensor. The image sensor also includes narrow-band pixels that generate narrow-band pixel signals that remain unread. The method also includes sending the wide-band pixel signals to an image signal processor, forming a partially-filled reference image based on data from the wide-band pixel signals; and applying an edge-forming technique to the partially-filled reference image to produce an edge map. An edge detection system includes a two-dimensional array of pixels having wide-band pixels and narrow-band pixels, an image signal processor for producing an edge map from a partially-filled reference image, and a readout circuit for generating the partially-filled reference image for the image signal processor. The partially-filled reference is based only on at least part of the wide-band pixels.