Attribute recognition method, system and equipment based on cross-domain auxiliary training and medium

The invention provides an attribute recognition method, system and device based on cross-domain auxiliary training and a medium. The method comprises the steps that a human body reference image is formed; performing human body part segmentation on each image in the image set without the attribute label to generate a human body part segmentation result, and randomly selecting a shielding picture; covering the shielding picture on the human body reference image to form a shielding composite image, and judging whether each part of the human body in the acquired human body reference image in the shielding composite image is visible or not; if not, correspondingly modifying the attribute label corresponding to the part, and forming an attribute label for shielding the composite image; and extracting each image feature, and inputting the image features to a preset attribute prediction model and a preset human body prediction model to obtain an attribute prediction result and a human body prediction result. Accordin