A visual search method for datasets identifies requests from client digital data devices that contain images, apparent objects of interest in the images, and bounding boxes in the images of those apparent objects. Includes using a detection model for. For each of one or more of the apparent objects of interest, the method extracts the sub-images defined by each bounding box. Feature search models are used to identify the features of the apparent object in each of those sub-images, and these features are sent to search engines (eg, text or text) to identify items in the digital dataset. (Otherwise) applies. The results of the search may be presented on the requesting user's digital data device.
展开▼