US 11,785,174 B2
Method for implanting information into video, computer device and storage medium
Chenqiong Gao, Shenzhen (CN); Zelong Yin, Shenzhen (CN); and Nianhua Xie, Shenzhen (CN)
Assigned to Tencent Technology (Shenzhen) Company Limited, Guangdong (CN)
Filed by Tencent Technology (Shenzhen) Company Limited, Guangdong (CN)
Filed on Aug. 5, 2021, as Appl. No. 17/394,579.
Application 17/394,579 is a continuation of application No. PCT/CN2020/085939, filed on Apr. 21, 2020.
Claims priority of application No. 201910385878.4 (CN), filed on May 9, 2019.
Prior Publication US 2021/0368112 A1, Nov. 25, 2021
Int. Cl. H04N 9/68 (2023.01); H04N 5/272 (2006.01)
CPC H04N 5/272 (2013.01) [H04N 9/68 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for embedding information into a video, applicable to an electronic device and comprising:
constructing a model for each of a plurality of pixels comprised by a first embedding region in a reference frame of the video, the model conforming to a pixel distribution characteristic of the first embedding region;
controlling update of the model based on a subsequent to-be-processed frame of the reference frame;
identifying a background and a foreground of a second embedding region in the to-be-processed frame based on the model;
generating a template used for occluding the background and exposing the foreground;
applying the template to to-be-embedded information to generate an updated to-be-embedded information by occluding region of the to-be-embedded information corresponding to the foreground, by steps comprising:
performing a multiplication operation on the pixels of the to-be-embedded information with a number filled in a corresponding position in the template; and
overlaying the second embedding region with the updated to-be-embedded information, so that the foreground of the second embedding region is devoid of occlusion by the overlaid to-be-embedded information.