Method for image generation, electronic device and non-transitory storage medium

An electronic device comprising:

• at least one processor;
• a memory communicatively connected with the at least one processor;
• wherein
• the memory is arranged to store instructions executable by the at least one processor, the instructions, when executed by the at least one processor, enabling the at least one processor to execute the following steps:
• obtaining a source image, a source prompt, a target prompt, a first weight and a second weight, wherein the source prompt is used for providing a source text condition corresponding to the source image, the target prompt is used for providing a target text condition to obtain a target image, the first weight is a weight of a guidance scale shared for inversion and generation in a source branch, the second weight is a weight of a guidance scale kept for generation in a target branch, the source branch is used for recovering the source image, and the target branch is used for generating the target image;
• performing inversion with the first weight on the source image to obtain an inversion result;
• determining a second noise based on a first noise, and determining a third noise based on the second noise, wherein the second noise is an initial noise for the source branch to preserve knowledge from the source image and the third noise is an initial noise for the target branch to obtain a new image;
• obtaining a recovered image based on the second noise, the source prompt and the first weight in the source branch;
• and
• obtaining the target image based on the recovered image, the third noise, the target prompt and the second weight in the target branch.

A non-transitory storage medium, wherein the non-transitory storage medium comprises:

• an executable program, and when the executable program runs, an electronic device on which the non-transitory storage medium is located to execute the following steps:
• the memory is arranged to store instructions executable by the at least one processor, the instructions, when executed by the at least one processor, enabling the at least one processor to execute the following steps:
• obtaining a source image, a source prompt, a target prompt, a first weight and a second weight, wherein the source prompt is used for providing a source text condition corresponding to the source image, the target prompt is used for providing a target text condition to obtain a target image, the first weight is a weight of a guidance scale shared for inversion and generation in a source branch, the second weight is a weight of a guidance scale kept for generation in a target branch, the source branch is used for recovering the source image, and the target branch is used for generating the target image;
• performing inversion with the first weight on the source image to obtain an inversion result;
• determining a second noise based on a first noise, and determining a third noise based on the second noise, wherein the second noise is an initial noise for the source branch to preserve knowledge from the source image and the third noise is an initial noise for the target branch to obtain a new image;
• obtaining a recovered image based on the second noise, the source prompt and the first weight in the source branch;
• and
• obtaining the target image based on the recovered image, the third noise, the target prompt and the second weight in the target branch.