US 12,087,070 B2
Sequence labeling task extraction from inked content
Jenna Hong, Acton, MA (US); Apurva Sandeep Gandhi, Union City, CA (US); Gilbert Antonius, San Ramon, CA (US); Tra My Nguyen, Brighton, MA (US); Ryan Serrao, Seattle, WA (US); Biyi Fang, Bellevue, WA (US); and Sheng Yi, Bellevue, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Nov. 12, 2021, as Appl. No. 17/454,729.
Prior Publication US 2023/0154218 A1, May 18, 2023
Int. Cl. G06F 40/284 (2020.01); G06V 10/22 (2022.01); G06V 30/32 (2022.01); G06V 30/412 (2022.01)
CPC G06V 30/36 (2022.01) [G06F 40/284 (2020.01); G06V 10/22 (2022.01); G06V 30/333 (2022.01); G06V 30/412 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer system comprising:
one or more processors configured to:
receive user input for inked content to a digital canvas;
process the inked content to determine one or more writing regions, each writing region including recognized text and one or more document layout features associated with that writing region;
tokenize a target writing region of the one or more writing regions into a sequence of tokens, the sequence of tokens including tokens representing recognized text and tokens representing the one or more document layout features;
process the sequence of tokens of the target writing region using a task extraction subsystem that operates on tokens representing both the recognized text and the one or more document layout features of the target writing region, the task extraction subsystem being configured to segment the target writing region into one or more sentence segments and classify each of the one or more sentence segments as a task sentence or a non-task sentence; and
extract one or more sentence segments that have been classified as task sentences.