公开(公告)号：US11450008B1

公开(公告)日：2022-09-20

申请号：US16803363

申请日：2020-02-27

Applicant: Amazon Technologies, Inc.

Inventor： Ambrish Tyagi ,  Siddhartha Chandra ,  Amit Kumar Agrawal ,  Viveka Kulharia

IPC: G06T7/194 ,  G06N3/04 ,  G06F17/18 ,  G06F17/16 ,  G06T11/20

Abstract: Devices and techniques are generally described for weakly-supervised object segmentation in image data. In various examples, a first frame of image data may be received. The first frame may include a first bounding box surrounding a first set of pixels, wherein first subset of pixels of the first set of pixels represent a first object of a first class and wherein second subset of pixels of the first set of pixels represent background image data. Cross-entropy loss may be determined for the first set of pixels. In some examples, a spatial attention map may be determined for the first set of pixels. In further examples, parameters of a convolutional neural network may be determined by modulating the cross-entropy loss for the first set of pixels using the spatial attention map. The convolutional neural network may be used to generate a segmentation map.