US 11,886,499 B2
Apparatus for training recognition model, apparatus for analyzing video, and apparatus for providing video search service
Jeong-Woo Son, Daejeon (KR); Chang-Uk Kwak, Daejeon (KR); Sun-Joong Kim, Sejong-si (KR); Alex Lee, Daejeon (KR); Min-Ho Han, Daejeon (KR); and Gyeong-June Hahm, Daejeon (KR)
Assigned to ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE, Daejeon (KR)
Filed by ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE, Daejeon (KR)
Filed on Feb. 3, 2021, as Appl. No. 17/166,444.
Claims priority of application No. 10-2020-0082871 (KR), filed on Jul. 6, 2020.
Prior Publication US 2022/0004773 A1, Jan. 6, 2022
Int. Cl. G06K 9/00 (2022.01); G06K 9/46 (2006.01); G06K 9/62 (2022.01); G06T 7/70 (2017.01); G06T 7/00 (2017.01); G11B 27/34 (2006.01); G11B 27/10 (2006.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); G06F 16/735 (2019.01); G06F 16/738 (2019.01); G06F 16/75 (2019.01); G06V 20/40 (2022.01); G06F 18/21 (2023.01); G06V 10/82 (2022.01)
CPC G06F 16/75 (2019.01) [G06F 16/735 (2019.01); G06F 16/738 (2019.01); G06F 18/21 (2023.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); G06T 7/0002 (2013.01); G06T 7/70 (2017.01); G06V 10/82 (2022.01); G06V 20/46 (2022.01); G06V 20/49 (2022.01); G11B 27/102 (2013.01); G11B 27/34 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30168 (2013.01); G06T 2207/30244 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An apparatus for training a recognition model, comprising:
at least one program and memory in which the program is recorded; and
a processor for executing the at least one program,
wherein the at least one program includes
a shot composition recognition model generation unit for generating a neural network model for predicting a shot composition and a camera position using a video shot tagged with shot composition information and camera position information as training data, and
a shot time and location recognition model generation unit for generating a neural network model for predicting a shot time and a shot location using a video shot tagged with shot time information and shot location information as training data,
wherein the shot time and location recognition model generation unit comprises a recognition-model training unit for training a shot location recognition model or a shot time recognition model to predict shot location information or shot time information with which the at least one frame is tagged when at least one of a shot composition of an extracted frame, a color distribution of the extracted frame, and a key frame is input.