US 12,437,208 B2
Picture searching method and apparatus, electronic device and computer readable storage medium
Yong Li, Beijing (CN); Yicheng Zhao, Beijing (CN); and Rui Li, Beijing (CN)
Assigned to BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD., Beijing (CN)
Filed by BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD., Beijing (CN)
Filed on Jan. 19, 2022, as Appl. No. 17/578,663.
Claims priority of application No. 202110136576.0 (CN), filed on Feb. 1, 2021.
Prior Publication US 2022/0245465 A1, Aug. 4, 2022
Int. Cl. G06N 5/01 (2023.01); G06F 16/532 (2019.01); G06F 16/535 (2019.01); G06F 16/538 (2019.01); G06F 16/55 (2019.01); G06F 16/583 (2019.01); G06F 18/243 (2023.01); G06N 5/04 (2023.01); G06N 5/048 (2023.01); G06N 20/00 (2019.01); G06V 10/77 (2022.01); G06V 10/771 (2022.01)
CPC G06N 5/01 (2023.01) [G06F 16/535 (2019.01); G06V 10/771 (2022.01); G06V 10/7715 (2022.01); G06F 16/532 (2019.01); G06F 16/538 (2019.01); G06F 16/55 (2019.01); G06F 16/583 (2019.01); G06F 18/24323 (2023.01); G06N 5/041 (2013.01); G06N 5/048 (2013.01); G06N 20/00 (2019.01); Y02D 10/00 (2018.01)] 17 Claims
OG exemplary drawing
 
1. A picture searching method, comprising:
acquiring features of pictures in a picture set to be searched, generating a decision tree according to the features, and configuring corresponding questions for nodes in the decision tree respectively; and
if a picture searching request of a user is received, asking the user the question, selecting an optimal path from the decision tree in conjunction with an answer of the user, and taking the picture corresponding to a leaf node on the optimal path as a searched picture,
wherein the selecting an optimal path from the decision tree comprises:
taking a root node in the decision tree as a current node, and performing the following first processing operation:
asking the user the question corresponding to the current node;
acquiring an answer of the user, and determining a child node matched with the answer from child nodes of the current node; and
if the matched child node is determined to be a leaf node, taking a path from the root node to the leaf node as the optimal path, otherwise, taking the matched child node as the current node, and repeatedly executing the first processing operation.