Method and mobile terminal for realizing video preview and retrieval

Abstract

The invention discloses a method for realizing video preview and retrieval, which comprises the following steps of: setting a buffer sequence with variable length; extracting a video file and processing the video file to obtain an image; detecting a scene lens in the obtained image and a key frame of the scene lens, and storing the detected scene lens and the detected key frame into the set buffer sequence with the variable length; and extracting the scene lens and the key frame from the buffer sequence with the variable length, matching and comparing the scene lens and the key frame with a specified target image, and performing the video retrieval. The invention also discloses a mobile terminal for realizing the video preview and the retrieval. According to the technical scheme of the invention, content preview and quick retrieval of the video file are realized in the mobile terminal with relatively limited processing capability and memory resource.

Description

technical field [0001] The invention relates to the field of video management in mobile terminals, in particular to a method for realizing video preview and retrieval and a mobile terminal. Background technique [0002] With the rapid development of integrated circuit technology, mobile terminals already have powerful processing capabilities, and mobile terminals are changing from simple communication tools to a comprehensive information processing platform. Modern mobile terminals already have extremely powerful processing capabilities, memory, solidified storage media, and computer-like operating systems. Such mobile terminals are a complete ultra-small computer system that can complete complex processing tasks. With the rapid increase of expandable memory capacity in mobile terminals, users can already store videos in most formats in mobile phones, palmtop computers and other mobile terminals, but the data volume of video files is very large, and users face longer duratio...

AI Technical Summary

Problems solved by technology

With the rapid increase of expandable memory capacity in mobile terminals, users can already store videos in most formats in mobile phones, palmtop computers and other mobile terminals, but the data volume of video files is very large, and users face longer duration Long video files, currently you can only know the content of the video after watching the video from beginning to end

[0003] Considering that the content of video files often has large redundancy, if the main scenes in the video file can be extracted, the user can choose the part of the video content that he is interested in to watch; or, if the user has obtained a certain Video image frame, if you want to quickly locate the position of this frame and start watching the video from this position, usually you can only find it by manually dragging the video progress. This method is not only inefficient, but also because the screen of mobile terminals such as mobile phones is relatively Small, the progress of dragging is difficult to grasp, so a slight jitter is likely to miss a lot of scenes you want to watch, so this method is difficult to achieve accurate positioning

[0004] In the prior art, the method for content retrieval of video files is: first capture the video frame, and then convert the captured video into many pictures. This step is generally implemented by using a third-party video decoder or provided by Microsoft. DirectX; in the process of capturing these pictures, compare the differences between the images of several adjacent frames, take the frames with large differences as key frames or use other more complex judgment methods, such as using the spatio-temporal attention model and so on to realize the acquisition of key frames; finally, a complex matching and comparison process is performed on the key frames and the target image to be retrieved, such as using expert systems, neural networks, etc. All the shots are obtained after all the shots. In this way, unplanned processing of all the shots of the video will take up too much memory, so this method is not suitable for mobile terminals

In addition, using this method, most of the complex analysis methods are used in the acquisition of key frames and matching judgments, and the amount of calculation is huge. Therefore, this method is only suitable for computers with increasingly powerful processing capabilities. For mobile terminals with relatively limited resources, the powerful resource consumption and calculation amount of this method are unbearable for mobile terminals

Images