Duc-Hieu TranFollow


Converting a handwritten note to a digital format requires a user to obtain an image of the note via a camera or scanner and use suitable techniques to convert the content within the captured image to a user-manipulable digital form. Editing the handwritten note after capture requires a repeat of the process to update the digital version. This is cumbersome and time consuming. This disclosure describes techniques that use a digital camera to automatically capture handwritten notes. The camera is positioned such that the entire notetaking surface is within the field of view. As the user writes, the camera captures the handwritten notes as a succession of images or a video stream. On-device machine learning models or other suitable techniques are used to convert the captured note to well-structured digital text in real time. Content from newly captured notes is compared to previously stored notes and if a match is found, the previous note is automatically updated, e.g., to remove erased content, to add new content, etc. Implementation of the described techniques can provide users with a seamless and efficient user experience (UX) of automatically capturing their handwritten notes in a structured digital form.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.