公开(公告)号：US10990877B2

公开(公告)日：2021-04-27

申请号：US15866129

申请日：2018-01-09

Applicant: Adobe Inc.

Inventor： Zhe Lin ,  Xiaohui Shen ,  Radomir Mech ,  Jian Ren

IPC: G06N3/08 ,  G06K9/62 ,  G06K9/00 ,  G06F16/783 ,  G06N3/04

Abstract: Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.

公开(公告)号：US11410038B2

公开(公告)日：2022-08-09

申请号：US17204370

申请日：2021-03-17

Applicant: Adobe Inc.

Inventor： Zhe Lin ,  Xiaohui Shen ,  Radomir Mech ,  Jian Ren

IPC: G06N3/08 ,  G06K9/62 ,  G06F16/783 ,  G06N3/04 ,  G06V20/40 ,  G06V40/16

Abstract: Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.

公开(公告)号：US20210201150A1

公开(公告)日：2021-07-01

申请号：US17204370

申请日：2021-03-17

Applicant: Adobe Inc.

Inventor： Zhe Lin ,  Xiaohui Shen ,  Radomir Mech ,  Jian Ren

IPC: G06N3/08 ,  G06K9/62 ,  G06K9/00 ,  G06F16/783 ,  G06N3/04

Abstract: Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.

公开(公告)号：US20190213474A1

公开(公告)日：2019-07-11

申请号：US15866129

申请日：2018-01-09

Applicant: Adobe Inc.

Inventor： Zhe Lin ,  Xiaohui Shen ,  Radomir Mech ,  Jian Ren

IPC: G06N3/08 ,  G06K9/62 ,  G06F17/30 ,  G06K9/00

CPC classification number: G06N3/08 ,  G06F16/784 ,  G06K9/00221 ,  G06K9/00744 ,  G06K9/6256

Abstract: Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.