公开(公告)号：US10713794B1

公开(公告)日：2020-07-14

申请号：US15922734

申请日：2018-03-15

Applicant: Facebook, Inc.

Inventor： Kaiming He ,  Georgia Gkioxari ,  Piotr Dollar ,  Ross Girshick

IPC: G06N3/04 ,  G06K9/32 ,  G06T7/11

Abstract: In one embodiment, a method includes a computing system accessing a training image. The system may generate a feature map for the training image using a first neural network. The system may identify a region of interest in the feature map and generate a regional feature map for the region of interest based on sampling locations defined by a sampling region. The sampling region and the region of interest may correspond to the same region in the feature map. The system may generate an instance segmentation mask associated with the region of interest by processing the regional feature map using a second neural network. The second neural network may be trained using the instance segmentation mask. Once trained, the second neural network is configured to generate instance segmentation masks for object instances depicted in images.

公开(公告)号：US20190156210A1

公开(公告)日：2019-05-23

申请号：US16192649

申请日：2018-11-15

Applicant: Facebook, Inc.

Inventor： Kaiming He ,  Ross Girshick ,  Xiaolong Wang

IPC: G06N3/08 ,  G06N3/04 ,  G06F17/15 ,  G06F17/30

Abstract: In one embodiment, a method includes training a baseline machine-learning model based on a neural network comprising a plurality of stages, wherein each stage comprises a plurality of neural blocks, accessing a plurality of training samples comprising a plurality of content objects, respectively, determining one or more non-local operations, wherein each non-local operation is based on one or more pairwise functions and one or more unary functions, generating one or more non-local blocks based on the plurality of training samples and the one or more non-local operations, determining a stage from the plurality of stages of the neural network, and training a non-local machine-learning model by inserting each of the one or more non-local blocks in between at least two of the plurality of neural blocks in the determined stage of the neural network.

公开(公告)号：US10984245B1

公开(公告)日：2021-04-20

申请号：US16286377

申请日：2019-02-26

Applicant: Facebook, Inc.

Inventor： Du Le Hong Tran ,  Kaiming He ,  Heng Wang ,  Matthew Dan Feiszli ,  Lorenzo Torresani

IPC: G06K9/00 ,  G06N3/08 ,  G06K9/62

Abstract: In one embodiment, a method includes receiving a request for information associated with a video, determining the information associated with the video by processing the video using a machine-learning model which is based on a convolutional neural network comprising a plurality of layers, wherein at least one of the plurality of layers comprises one or more building blocks, wherein at least one of the one or more building blocks comprises a first filter configured to perform a three-dimensional (3D) pointwise convolutional operation and a second filter configured to perform a three-dimensional (3D) groupwise convolutional operation, and outputting the information associated with the video in response to the request.