Context-based shape extraction and interpretation from hand-drawn ink input

摘要

The electronic devices described herein are configured to enhance user experience associated with drawing or otherwise inputting shape data into the electronic devices. Shape input data is identified and matched against known shape patterns and, when a match is found, an entity associated with the shape is determined. The entity is converted into an annotation for rendering and/or displaying to the user. The shape identification, entity determination, and annotation conversion may all be based on one or more context elements to increase the accuracy of the shape interpretation. In particular, elements of conversations held via the electronic devices may be used as context for the shape interpretation. Further, machine learning techniques may be applied based on a variety of feedback data to improve the accuracy, speed, and/or performance of the shape interpretation process.