公开(公告)号：US12120450B2

公开(公告)日：2024-10-15

申请号：US17780586

申请日：2020-11-27

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Bin Xiao ,  Huanwen Peng ,  Nan Song ,  Yuanwen Liu ,  Yihua Zeng ,  Congchao Zhu

IPC: H04N5/262 ,  H04N5/272 ,  H04N5/91 ,  H04N23/63 ,  H04N23/667

CPC classification number: H04N5/2621 ,  H04N5/272 ,  H04N5/91 ,  H04N23/631 ,  H04N23/667

Abstract: A photographing method includes: starting a camera in response to a first operation; entering a first video recording mode in response to a second operation; capturing first video frames in response to a third operation; and stopping capturing the first video frames in response to a fourth operation, to obtain a target video. The ith second video frame in the target video is obtained by fusing a foreground image of the kth first video frame captured by the camera and a background image of the jth first video frame captured by the camera, and the (i+1)th second video frame in the target video is obtained by fusing a foreground image of the (k+m)th first video frame captured by the camera and a background image of the (j+n)th first video frame captured by the camera, where m≠n. In this way, the electronic device can record a video with a special effect.

公开(公告)号：US20160323585A1

公开(公告)日：2016-11-03

申请号：US15204725

申请日：2016-07-07

Applicant: Huawei Technologies Co., Ltd.

Inventor： Haichang Zhai ,  Yuanwen Liu ,  Changcai Lai ,  Yang Song

IPC: H04N19/159 ,  H04N19/187 ,  H04N19/119 ,  H04N19/11

CPC classification number: H04N19/159 ,  H04N19/11 ,  H04N19/119 ,  H04N19/147 ,  H04N19/176 ,  H04N19/187 ,  H04N19/50 ,  H04N19/61 ,  H04N19/70

Abstract: Embodiments of the present invention provide a video coding method for INTRA and a video coding apparatus. The method includes performing a traversal prediction on a current-layer coding unit CU, a prediction mode of a sub-Cu that is divided, from the current-layer CU, and a syntax element that is needed for coding of the current-layer CU; and if the best CU division mode is to divide the current-layer CU into four sub-CUs whose depth is N+1, and prediction modes of the four sub-CUs whose depth is N+1 are the same, performing coding, by using the current-layer CU as a coding unit, on the current-layer CU according to the best CU division mode, the prediction mode of the sub-CU that is divided from the current-layer CU according to the best CU division mode, and the syntax element that is needed for coding of the current-layer CU.

Abstract translation: 本发明的实施例提供了一种用于INTRA和视频编码装置的视频编码方法。 该方法包括对当前层编码单元CU执行遍历预测，从当前层CU划分的子Cu的预测模式和当前层CU编码所需的语法元素 ; 并且如果最佳CU分割模式是将当前层CU划分为深度为N + 1的四个子CU，并且深度为N + 1的四个子CU的预测模式相同，则通过 使用当前层CU作为编码单元，根据最佳CU分割模式在当前层CU上根据最佳CU分割模式从当前层CU划分的子CU的预测模式， 以及编码当前层CU所需的语法元素。

公开(公告)号：US12096134B2

公开(公告)日：2024-09-17

申请号：US17484093

申请日：2021-09-24

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Xijia Deng ,  Yuanwen Liu

IPC: H04N23/959 ,  G06N3/045 ,  G06T5/50 ,  G06T5/70 ,  H04N23/63

CPC classification number: H04N23/959 ,  G06N3/045 ,  G06T5/50 ,  G06T5/70 ,  H04N23/635

Abstract: This application provides an image processing method, including: after binocular image data of a photographing scene that is generated by dual cameras is obtained, determining binocular depth data by using the binocular image data, to determine an error region in an image, correcting depths of the error region based on time of flight (TOF) data corresponding to the scene, and finally performing blurring processing on monocular image data in the binocular image data by using corrected binocular depth data. Because TOF is considered when the depths of the error region are corrected, the method can effectively avoid a depth estimation error caused by relatively poor precision and stability of a binocular depth estimation manner, so that depth estimation precision and stability are improved.