Face abstracting method and video abstracting method based on face recognition and devices thereof

Abstract

The invention discloses a face abstracting method based on face recognition. The face abstracting method comprises the steps of generating face images of different persons appearing in an original video and forming an arisen face image list. The method further comprises the steps of face detection, face feature extraction, face feature clustering, face abstract image generation and the like, wherein the face detection is used for scanning image frames in the original video and judging whether face areas exist in the video frames. The invention further discloses a device for implementing the face abstracting method. According to the face abstracting method based on face recognition and the device thereof, the face images of different persons appearing in the video can be generated, the generated face image of each person is the most representative image appearing in the video and the corresponding position information in the original video. The face images can be rapidly recognized manually, and can also be used as input data of a current face identity recognition system or a face identity recognition system with the higher precision in the future.

Description

technical field [0001] The present invention relates to an automatic information providing technology, in particular to a method for information inquiry, matching and feedback by means of communication technology, and a device for realizing the method, so as to realize timely self-matching of umbilical cord blood stem cell supply and demand, so as to benefit patients. Get timely medical treatment. Background technique [0002] As one of the most important aspects in the construction of safe cities and smart cities, video capture terminals have been widely deployed. However, a large amount of video data has caused various problems in storage, viewing and retrieval. In the process of video reconnaissance, it is necessary to view a large number of videos and find the target within a limited time, which is a very difficult task. On the one hand, a large amount of video data requires a lot of manual inspection, on the other hand, manual inspection may miss potential targets. Ma...

AI Technical Summary

Problems solved by technology

[0004] However, video summarization and retrieval techniques are often only suitable for video scenes with sparse objects. There are two main reasons. One is that due to the limitations of computer vision technology itself, it is difficult for scenes with densely superimposed objects to be solved under the condition of limited computing resources. Accurate target segmentation and feature extraction are achieved. On the other hand, for scenes with dense targets, the comparison between the result video of the video summary and the original video does not significantly improve the viewing efficiency.

Therefore, in the case of dense target scenes, video summarization and retrieval techniques are often difficult to be practically applied

Although face recognition technology has made great progress, in some defined standard data sets, although the method of deep learning has reached or even exceeded the human recognition ability, it can be actually applied to the identification and verification of some ID photos, etc. However, in the actual surveillance video, due to the influence of various factors that are difficult to control, such as angle, illumination and distance, accurate face recognition is far from meeting people's expectations

[0005] The information of "people" in the video is often the main target that is looked for in the video. Other types of targets contain less information, but consume a lot of computing resources and time in video structural processing.

Images