| 发明名称 |
Method and apparatus for recognizing image content |
| 摘要 |
Embodiments of the present invention provide a method and apparatus for recognizing image contents. In one embodiment of the present invention, there is provided a method for recognizing image contents, comprising: providing at least a first image pair and a second image pair to a user; obtaining a first answer and a second answer from the user, wherein the first answer and the second answer respectively indicate the user's determination on whether the image contents in the first image pair and the second image pair are same or not; and modifying a weight of the second image pair in response to the first answer matching a predetermined answer. In one embodiment of the present invention, there is provided an apparatus for recognizing image contents. |
| 申请公布号 |
US9436930(B2) |
申请公布日期 |
2016.09.06 |
| 申请号 |
US201314089909 |
申请日期 |
2013.11.26 |
| 申请人 |
EMC Corporation |
发明人 |
Zhong Qianjie;Guo Feng;Chen Qiyan;Chen Chao;Chen Bo |
| 分类号 |
G06F21/36;H04L9/32;G06Q10/10 |
主分类号 |
G06F21/36 |
| 代理机构 |
|
代理人 |
Gupta Krishnendu;Lee Konrad R. |
| 主权项 |
1. A method-for recognizing-images, wherein the images are stored within a database which includes pairs of images that are same or unknown, comprising:
storing image pairs within the database; providing, using one or more processors, at least a first image pair and a second image pair; wherein the first image pair is selected from a first database that stores image pairs, in which whether the image contents are same or not is known; and the second image pair is selected from a second database that stores image pairs, in which whether the image contents are same or not is unknown; obtaining, using one or more processors, a first answer and a second answer from the user, wherein the first answer and the second answer respectively indicate the user's determination on whether the image contents in the first image pair and the second image pair are same or not; determining, using one or more processors, the image contents in the second image pair being same in response to the weight of the second image pair reaching a predetermined weight; modifying a weight of the second image pair in response to the first answer matching a predetermined answer; and wherein the verification of each pair of images is enabled to be changed based on the modifying of the weight. |
| 地址 |
Hopkinton MA US |
