Abstract: The disclosure discloses a method, an apparatus, a device, and a storage medium for interaction. The method of interaction includes: in response to detecting that a first user enters a virtual room, displaying a shooting task list in the virtual room; the shooting task list containing a plurality of shooting tasks, and each shooting task carries task information; obtaining, for the shooting task, an image taken by the first user based on the task information carried by the shooting task; performing a feature extraction on the image to obtain feature information; and comparing the feature information with the task information, and determining that the shooting task is completed if the feature information matches the task information.

Abstract: Aspects of this disclosure include technologies for character-based text detection and recognition. The disclosed single-stage model is configured for joint text detection and word recognition in natural images. In the disclosed solution, a character recognition branch is integrated into a word detection model. This results in an end-to-end trainable model that can implement text detection and word recognition jointly. Further, the disclosed technical solution includes an iterative character detection method, which is configured to generate character-level bounding boxes on real-world images by using synthetic data first.