Home > News content

Google is pouring more information from the knowledge map into image search

via:cnBeta.COM     time:2020/7/9 9:01:24     readed:75

According to foreign media,Google announced on Wednesday that it was upgrading image search results on U.S. mobile phones to gather more information from its knowledge graph.From the billions of pieces of information collected in google's knowledge map database, information about people, places, or objects associated with pictures is now displayed below the picture, giving users more information about the picture and its surroundings.

visit:

Alibaba cloud new user welfare special session ECS as low as 102 yuan / year

Day Wing Cloud Year Cloud Day Cloud Festival Cloud Host 1 C2G 92 Yuan / year Real Name Registration 8888 Yuan Gift Pack

`SDY5~038HNPE_XDIFC~X5M.png

For example, suppose a user is searching for a beautiful state park nearby and wants to visit it. Users want to swim during the tour, so they may click on a picture of a park with a river. At the bottom of the photo, users may see related themes, such as the name of the river, or which city the park is in. If a user clicks on a specific topic, it will expand and show the user a short description of the person, place or thing it refers to, as well as a link to learn more and other related topics for users to explore.

ezgif-5-4302551d9c5e.gif

to generate links to relevant knowledge map entities, google through deep learning, evaluates their understanding of images, evaluates visual and text signals of images, and combines with google's understanding of text on image pages. This information helps them identify the most likely people, places, or things associated with a particular image. They match it with existing topics in the knowledge map and then emerge in Google Pictures when they are sure to find a matching picture.

2.gif

China IT News APP

Download China IT News APP

Please rate this news

The average score will be displayed after you score.

Post comment

Do not see clearly? Click for a new code.

User comments