| 摘要 |
<p>The present invention relates to a method of retrieving a face in a digital image, characterized by the steps of: detecting a face in a scene of an image; extracting the features out of the face; building a feature vector for a specific number of faces to be sent; sending the features to the server using SOAP/REST technology (SET Request); verifying the process by extracting the feature of each face; building the result feature for each face, which contains different results, ranked by least dissimilarity; and sending back the result vector for each face (GET Request) to the client; whereby the results of the matching face will be sent as RESET request; whilst the client side will use the similarity values embedded in the vector to make the decision on the outcome value.</p> |
