Image processing device, image display system, image processing method, and recording medium

An image processing method comprising:

• obtaining an image which has been captured from a movable body;
• estimating levels of importance with respect to a respective plurality of regions included in the image by an importance level estimating model which has been trained;
• estimating preliminary quality parameters with respect to the respective plurality of regions included in the image;
• determining quality parameters with respect to the respective plurality of regions with reference to the preliminary quality parameters which have been estimated and the levels of importance;
• generating replaced data by replacing at least one of the plurality of regions, which are included in the image, with alternative data in accordance with the levels of importance;
• deriving an evaluation value by referring to the replaced data;
• and
• training the importance level estimating model with reference to the evaluation value.

A non-transitory recording medium in which a program for causing a computer to function as an information processing apparatus is stored, the program causing the computer to execute operations comprising:

• an obtaining process comprising obtaining an image which has been captured from a movable body;
• an importance level estimating process comprising estimating levels of importance with respect to a respective plurality of regions included in the image by an importance level estimating model which has been trained;
• a quality parameter estimating process comprising estimating preliminary quality parameters with respect to the respective plurality of regions included in the image;
• a quality parameter determining process comprising determining quality parameters with respect to the respective plurality of regions with reference to the preliminary quality parameters which have been estimated and the levels of importance;
• a replacing process comprising generating replaced data by replacing at least one of the plurality of regions, which are included in the image, with alternative data in accordance with the levels of importance;
• an evaluating process comprising deriving an evaluation value by referring to the replaced data;
• and
• a training process comprising training the importance level estimating model with reference to the evaluation value.