Leveraging Previous Instances of Handwriting for Handwriting Beautification and Other Applications

摘要

A stroke processing system (SPS) is described which processes a user's current instance of handwriting with reference to previous instances of handwriting. In one application, the SPS matches a new token input by a user (corresponding to a sequence of stroke samples) with previous tokens that are similar to the new token. The SPS then refines the new token by averaging it with the previous tokens. This improves the appearance of the new token. In another application, the SPS uses the previous tokens to produce a search result. In another application, the SPS uses the previous tokens to perform an auto-completion operation.

主权项

1. A stroke processing system, implemented by one or more computer devices, for processing handwriting, comprising:
a token formation module configured to:
receive input stroke samples that represent handwriting produced using an input device;formulate a new token based on the input stroke samples; a data store that stores a collection of previous tokens; a similarity assessment module configured to examine the collection of previous tokens to determine if there is at least one previous token that is similar to the new token; and a token refinement module configured to modify an appearance of the new token based on said at least one previous token.