-
公开(公告)号:US20240176959A1
公开(公告)日:2024-05-30
申请号:US18200778
申请日:2023-05-23
Inventor: Jeong Heo , Young-Ae SEO , Jin SEONG , Jong Hun SHIN , Ki Young Lee , Soojong LIM , Young Kil Kim , Jihee Ryu
Abstract: Provided is a method of generating a language model using crossmodal information. The method includes: receiving language-based first modality information and non-language-based second modality information; converting the first modality information into a first byte sequence; converting the second modality information into a second byte sequence; converting the first and second byte sequences into a first embedding vector and a second embedding vector by applying an embedding technique for each modality; generating semantic association information between first and second modality information by inputting the first and second embedding vectors to a crossmodal transformer; and learning the language model by setting the generated semantic association information as training data.
-
2.
公开(公告)号:US20150009324A1
公开(公告)日:2015-01-08
申请号:US14321768
申请日:2014-07-01
Inventor: Hyung Kyu LEE , Soojong LIM , Song-woo SOK , Ok Gee MIN , Kyoung PARK
IPC: H04N7/18
CPC classification number: H04N7/181 , G01S13/72 , G01S13/867 , G08B13/19608 , G08B13/19645 , G08B13/19695 , G08B13/19697
Abstract: An image monitoring apparatus is provided to monitor a wide area and accurately identify an object of surveillance based on radar technology. The monitoring area of a surveillance camera can be extended using radar, and an object to be detected can be accurately identified using fixed background information.
Abstract translation: 提供一种图像监视装置,用于监视广域,并基于雷达技术准确识别监视对象。 可以使用雷达来扩展监视摄像机的监视区域,并且可以使用固定的背景信息来准确地识别要检测的对象。
-