US 12,294,773 B2
Method and apparatus for generating video digest and readable storage medium
Chengtao Zhou, Guangdong (CN); Le Zhang, Guangdong (CN); and Zuoxing Yang, Guangdong (CN)
Assigned to SHENZHEN MICROBT ELECTRONICS TECHNOLOGY CO., LTD., Shenzhen (CN)
Appl. No. 18/710,957
Filed by SHENZHEN MICROBT ELECTRONICS TECHNOLOGY CO., LTD., Guangdong (CN)
PCT Filed Apr. 24, 2023, PCT No. PCT/CN2023/090361
§ 371(c)(1), (2) Date May 16, 2024,
PCT Pub. No. WO2023/231647, PCT Pub. Date Dec. 7, 2023.
Claims priority of application No. 202210614997.4 (CN), filed on Jun. 1, 2022.
Prior Publication US 2024/0334027 A1, Oct. 3, 2024
Int. Cl. H04N 21/8549 (2011.01); H04N 21/44 (2011.01)
CPC H04N 21/8549 (2013.01) [H04N 21/44008 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A method for generating a video digest, wherein the method is applied to a video acquisition device, the video acquisition device comprises a hardware unit configured to obtain movement information in real time, and the movement information comprises speed information and direction information of the video acquisition device, the method comprising:
starting acquisition of a video signal in response to an acquisition start instruction, and recording movement information corresponding to each frame during the acquisition of the video signal;
determining a frequency for key frame extraction according to the movement information corresponding to each frame;
wherein when the movement information comprises the speed information describing a speed at which the video acquisition device is moving, the frequency for key frame extraction is proportional to the speed information; and
wherein when the movement information comprises direction information describing a direction in which the video acquisition device is moving and head turning and/or steering is determined to have occurred according to a change of the direction information, the frequency for key frame extraction is increased;
determining a key frame from the acquired video signal according to the frequency for key frame extraction without analyzing frames of the video signal to extract video features; and
in response to an acquisition end instruction, generating a video digest according to the key frame.