An image processing apparatus includes an imaging unit configured to capture an image, an extraction unit configured to extract a feature amount in each image region of a plurality of objects from the captured image, a determination unit configured to determine, based on the extracted feature amount, a type of a feature amount used for discriminating each of the plurality of objects and a resolution of an image used for extracting a feature amount of the determined type, an imaging control unit configured to control the imaging unit to output an image at the determined resolution, and an extraction control unit configured to control the extraction unit to extract a feature amount of the determined type from the image output at the determined resolution from the imaging unit.