US 12,340,561 B2
Video retrieval system using object contextualization
Kwang Hoon Kim, Suwon-si (KR); and Dinh Lam Pham, Suwon-si (KR)
Assigned to KYONGGI UNIVERSITY INDUSTRY AND ACADEMIA COOPERATION FOUNDATION, Seongnam-si (KR)
Filed by KYONGGI UNIVERSITY INDUSTRY AND ACADEMIA COOPERATION FOUNDATION, Seongnam-si (KR)
Filed on Dec. 22, 2022, as Appl. No. 18/086,660.
Prior Publication US 2024/0212324 A1, Jun. 27, 2024
Int. Cl. G06K 9/00 (2022.01); G06F 16/732 (2019.01); G06F 16/783 (2019.01); G06V 10/70 (2022.01); G06V 20/40 (2022.01)
CPC G06V 10/768 (2022.01) [G06F 16/7335 (2019.01); G06F 16/7837 (2019.01); G06V 20/41 (2022.01); G06V 20/46 (2022.01)] 11 Claims
OG exemplary drawing
 
1. A video search system using object contextualization, comprising:
an object contextualization server configured to detect an object for each frame from an input video file, generate frame contextualization data including types and the number of objects detected for each frame, object attribute information, and location information of the object, and combine the frame contextualization data to generate object contextualization data for video data and store the generated object contextualization data as a text file in a big data storage; and
a video search server configured to receive a search condition and search for a video that meets the search condition based on text search with respect to the object contextualization data.