Image lighting methods and apparatuses, electronic devices, and storage media

An electronic device, comprising:

• a memory, configured to store computer-readable instructions;
• and
• a processor, configured to invoke and execute the computer-readable instructions in the memory to execute the following operations:
• determining a convolutional neural network corresponding to a lighting operation type of an initial image;
• obtaining local feature information and global feature information of the initial image by means of the convolutional neural network;
• obtaining, based on the local feature information and the global feature information, fusion feature information of the initial image by means of the convolutional neural network;
• obtaining, based on a luminance component map of the initial image, a maximum pooling result map and a minimum pooling result map by means of the convolutional neural network;
• obtaining, based on the fusion feature information, the maximum pooling result map, and the minimum pooling result map, a bilateral grid matrix of the initial image;
• and
• performing affine transformation on the initial image according to the bilateral grid matrix to obtain a target image, the target image being an image obtained after lighting the initial image according to the lighting operation type.

A non-transitory storage medium, configured to store computer-readable instructions, wherein execution of the instructions by the processor causes the processor to perform the following operations:

• determining a convolutional neural network corresponding to a lighting operation type of an initial image;
• obtaining local feature information and global feature information of the initial image by means of the convolutional neural network;
• obtaining, based on the local feature information and the global feature information, fusion feature information of the initial image by means of the convolutional neural network;
• obtaining, based on a luminance component map of the initial image, a maximum pooling result map and a minimum pooling result map by means of the convolutional neural network;
• obtaining, based on the fusion feature information, the maximum pooling result map, and the minimum pooling result map, a bilateral grid matrix of the initial image;
• and
• performing affine transformation on the initial image according to the bilateral grid matrix to obtain a target image, the target image being an image obtained after lighting the initial image according to the lighting operation type.