Watermark embedding method and apparatus, watermark extraction method and apparatus, and electronic device and storage medium

A watermark extraction method, comprising:

• obtaining a target video with a to-be-extracted watermark;
• according to a third quantity and a fourth quantity, dividing the target video to obtain respective to-be-extracted frame sequences corresponding to the target video, wherein the third quantity is a quantity of original videos used when a target video watermark is embedded in the target video, the fourth quantity is a quantity of video frames corresponding to a watermark code in the target video watermark, the to-be-extracted frame sequence comprises a third quantity of to-be-extracted watermark units, each of the to-be-extracted watermark units comprises a third quantity of to-be-extracted sub-units, and each of the to-be-extracted sub-units comprises a fourth quantity of consecutive video frames;
• for any one of the to-be-extracted sub-units in any one of the to-be-extracted watermark units, according to a pixel value of each of the video frames in the to-be-extracted sub-unit, determining a sub-unit watermark corresponding to the to-be-extracted sub-unit;
• and
• extracting respective target video watermarks corresponding to the target video according to a unit serial number of each of the to-be-extracted watermark units in each of the to-be-extracted frame sequences, the sub-unit watermark corresponding to each of the to-be-extracted sub-units, and a watermark type.

An electronic device, comprising:

• a processor, and a memory, wherein the memory stores machine-readable instructions executable by the processor, the processor is configured to execute the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the processor, the processor is caused to:
• obtain a plurality of original videos corresponding to a same video content, and a video serial number and a to-be-embedded watermark corresponding to each of the plurality of original videos, wherein the to-be-embedded watermark comprises at least one watermark code;
• for each original video of the plurality of original videos, divide the original video according to a first quantity and a second quantity to obtain a frame sequence of the original video, wherein the first quantity indicates a quantity of the plurality of original videos, the second quantity indicates a quantity of video frames corresponding to the at least one watermark code in the to-be-embedded watermark, the frame sequence comprises the first quantity of watermark units, each of the watermark units comprises the first quantity of sub-units, and each of the sub-units comprises the second quantity of consecutive video frames;
• allocate a watermark code for each of the frame sequences according to a timing of each of the frame sequences and an order of each watermark code, and determine a first watermark and a second watermark corresponding to each of the frame sequences according to the watermark code allocated to each of the frame sequences;
• and
• add a target watermark for each sub-unit to obtain a watermark video corresponding to the original video, according to the video serial number of each of the plurality of original videos, a first serial number of each watermark unit in each of the frame sequences, and a second serial number of the sub-unit in each watermark unit, wherein the target watermark is one of a group consisting of the first watermark, the second watermark and the watermark code.