| CPC G06V 10/25 (2022.01) [G06N 3/045 (2023.01); G06V 10/776 (2022.01); G06V 10/82 (2022.01)] | 10 Claims |

|
1. An image processing apparatus comprising:
at least one processor or circuit configured to function as:
a first foreground region detection unit configured to detect, based on a first neural network, a first foreground region in an inputted image;
a partial region identification unit configured to identify, based on a second neural network, a partial region in the inputted image in which a foreground region is detected by a method other than a neural network;
a second foreground region detection unit configured to detect, based on a detection result by a method other than a neural network, a second foreground region in the partial region; and
an output unit configured to output, based on the first foreground region, the partial region, and the second foreground region, a third foreground region indicating a foreground region in the inputted image,
wherein the second neural network performs learning based on a detection error of the first neural network.
|