Video feature learning method, device, electronic device and readable storage medium

Abstract

The embodiment of the invention provides a video feature learning method and device, an electronic device and a readable storage medium. The method comprises the steps that a to-be-trained video sample is acquired, wherein the video sample comprises a plurality of frames of images; the video sample is segmented to obtain a plurality of continuous video segments; visual features of each video segment are extracted, and the number of motion elements of each video segment is calculated; visual features of the video sample are extracted, and the number of motion elements of the video sample is calculated; based on the number of the motion elements of each video segment, the number of the motion elements of the video sample and preset constraint conditions, a target classification model is trained so as to obtain a trained target classification model, so that learning of the video features is achieved. Therefore, unsupervised learning of the video features can be achieved without knowing tags and classification information of a video, the resource and cost consumption can be reduced, and the method can be applied to a wide range of video scenes.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a video feature learning method, device, electronic equipment and readable storage medium. Background technique [0002] Video feature learning has a wide range of applications, such as video classification, similar video retrieval, video matching, etc. The current video feature learning method is mainly based on video tags and classification information, and the above video tags and classification information require manual labeling work, which consumes a lot of resources and costs in actual business application scenarios with a huge amount of data. Contents of the invention [0003] In order to overcome the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a video feature learning method, device, electronic equipment and readable storage medium, which can realize unsupervised learning of video features without knowing ...

AI Technical Summary

Problems solved by technology

The current video feature learning method is mainly based on video tags and classification information, and the above video tags and classification information require manual labeling work, which consumes a lot of resources and costs in actual business application scenarios with a huge amount of data

Images