IP Library Granted Patent US 12,566,795
Granted Patent B2
US 12,566,795 · App. 18/901,269 · Granted Mar 3, 2026

Method and system for synchronized search and retrieval of visual work instructions using artificial intelligence

Inventors: Sung Bum Park (Yongin-si, KR); Suehyun Chang (Seoul, KR)
Assignees: HOSEO UNIVERSITY ACADEMIC COOPERATION FOUNDATION; LIVINAI INC.
G06F16/71G06F16/178G06F16/7844G06V10/86G06V20/41G06V20/49G06V20/62
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,566,795
App. No.
18/901,269
Granted
Mar 3, 2026
Kind
B2
Abstract

The present invention relates to generating training data for performing artificial intelligence. A method and apparatus for generating section data of a visual work instruction for performing artificial intelligence are provided for generating data for searching a user's desired section in a visual work instruction using an artificial intelligence-based text search model.

Claims (19)

1 . A method for providing a search for a segment that a user is looking for in a video work instruction, comprising:

(a) separating a video segment from text information associated with the video work instruction, wherein the text information has a hierarchical structure including a module name, a unit name belonging to the module name, and a part name belonging to the unit name, and wherein the separating is performed based on a detection of a change in at least one of the unit name or the part name within the text information;

(b) generating and storing a text file corresponding to the video segment separated in step (a) based on the text information, wherein the generated text file integrates all of a description of an entire video work instruction, a task name, a task description, and the module name, the unit name, and the part name corresponding to the separated video segment into a single document for the separated video segment, and wherein, when the separating is performed based on a change in the unit name, the generated text file includes all part names associated with the unit name, whereas when the separating is performed based on a change in the part name, the generated text file includes a specific part name corresponding to the separated video segment;

(c) generating and storing synchronization information for the video segment separated in step (a) and the text file generated in step (b);

(d) searching for a text file corresponding to a query input by the user from among the stored text files;

(e) determining a video segment corresponding to the query input by the user by referring to the synchronization information of the searched text file; and

(f) providing the video segment corresponding to the query input by the user, determined in step (e), to the user,

wherein the synchronization information of step (c) is a start time and an end time of the video content for the text file generated in step (b).

2 . An apparatus for providing a search for a segment that a user is looking for in a video work instruction, comprising:

at least one processor; and

at least one memory for storing computer-executable instructions,

wherein the computer-executable instructions stored in the at least one memory cause the at least one processor to perform:

(a) separating a video segment from text information associated with the video work instruction, wherein the text information has a hierarchical structure including a module name, a unit name belonging to the module name, and a part name belonging to the unit name, and wherein the separating is performed based on a detection of a change in at least one of the unit name or the part name within the text information;

(b) generating and storing a text file corresponding to the video segment separated in step (a) based on the text information, wherein the generated text file integrates all of a description of an entire video work instruction, a task name, a task description, and the module name, the unit name, and the part name corresponding to the separated video segment into a single document for the separated video segment, and wherein, when the separating is performed based on a change in the unit name, the generated text file includes all part names associated with the unit name, whereas when the separating is performed based on a change in the part name, the generated text file includes a specific part name corresponding to the separated video segment;

(c) generating and storing synchronization information for the video segment separated in step (a) and the text file generated in step (b);

(d) searching for a text file corresponding to a query input by the user from among the stored text files;

(e) determining a video segment corresponding to the query input by the user by referring to the synchronization information of the searched text file; and

(f) providing the video segment corresponding to the query input by the user, determined in step (e), to the user,

wherein the synchronization information of step (c) is a start time and an end time of the video content for the text file generated in step (b).

Priority Claims (1)
KR 10-2022-0097120 · Aug 4, 2022 · national
Continuity (2)
Continuation 18230220 · Aug 4, 2023
Related Publication 20250021600A1 · Jan 16, 2025
References Cited (2)
US 11017488B2 · Evans · 2021 [cited by examiner]
US 20210064648A1 · Farri · 2021 [cited by examiner]