Beijing Baidu Netcom Science and Technology Co., Ltd.

Mediatek Corporation

Beijing Baidu Netcom Science Technology Co., Ltd

Le Shi

Wenpeng Ding

Xu Zhang

Jian-Liang Lin

Hung-Chih Lin

Shen-Kai Chang

Ya-Hsuan Lee

Chia-Ying Li

Provided are a video frame processing method, a model training method, a device and a storage medium, relates to a field of artificial intelligence, and in particular, to cloud computing, video processing, and medium cloud technology, and may be applied in an intelligent cloud scenario. The video frame processing method includes: acquiring a target characteristic corresponding to a current video frame in a video frame sequence to be encoded, in the case of the video frame sequence to be encoded satisfies a preset condition; inputting the target characteristic corresponding to the current video frame to a first target model, to obtain a first output result corresponding to the current video frame; and determining a first target group of pictures (GOP) length corresponding to the current video frame, based on the first output result corresponding to the current video frame.

Method for processing video frame, method for training model, electronic device, and storage medium

A method, apparatus, device and medium for coding a video are provided. An implementation plan includes: determining an initial video frame structure based on a preset threshold for a B-frame number; in response to determining that the initial video frame structure meets a preset condition, shortening the initial video frame structure to obtain a candidate video frame structure set; determining a target video frame structure from the candidate video frame structure set; and performing video coding on video frames in a to-be-coded video frame sequence according to the target video frame structure.

Method and device for coding video using optimal video frame structure, and storage medium

A method and apparatus for coding a video, device and medium are provided. An implementation of the method include: determining a first video frame structure and a second video frame structure based on a pre-set threshold for a B-frame number; determining a target video frame structure based on the first video frame structure, the second video frame structure, and a pre-set condition; and coding video frames in a to-be-coded video frame sequence according to the target video frame structure.

Method and apparatus for coding video, device and medium

A method and apparatus for coding a video, an electronic device and a computer-readable storage medium. A specific implementation scheme is: coding a sampled video frame sequence of a source video by using a preset coding mode to obtain a sample coding overhead; analyzing frame types of frames to be coded in the source video to obtain a frame type sequence of the frames to be coded; determining a predictive coding overhead of the source video according to the sample coding overhead of the source video, the frame type sequence of the frames to be coded, and quantization parameters of coding modes to be used corresponding to different frame types; and determining a coding strategy based on the predictive coding overhead and requirements of transmission bit rate limit information.

Method and apparatus for coding video, electronic device and computer-readable storage medium

A video processing method includes: receiving an omnidirectional image/video content corresponding to a viewing sphere, generating a sequence of projection-based frames according to the omnidirectional image/video content and a viewport-based cube projection layout, and encoding the sequence of projection-based frames to generate a bitstream. Each projection-based frame has a 360-degree image/video content represented by rectangular projection faces packed in the viewport-based cube projection layout. The rectangular projection faces include a first rectangular projection face, a second rectangular projection face, a third rectangular projection face, a fourth rectangular projection face, a fifth rectangular projection face, and a sixth rectangular projection face split into partial rectangular projection faces. The first rectangular projection face corresponds to user's viewport, and is enclosed by a surrounding area composed of the second rectangular projection face, the third rectangular projection face, the fourth rectangular projection face, the fifth rectangular projection face, and the partial rectangular projection faces.

Method and apparatus for generating and encoding projection-based frame with 360-degree content represented by rectangular projection faces packed in viewport-based cube projection layout

Growing community of inventors

Beijing, China

Le Shi