Video material marking method and device, equipment and medium
A video and material technology, applied in the field of video processing, can solve problems such as poor generalization performance of text labels, and achieve the effect of generalization performance improvement
Pending Publication Date: 2022-05-27
宁波深擎信息科技有限公司 +1
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, in the process of realizing the present invention, the inventors found that the above-mentioned existing video description tasks still have the technical problem of poor generalization performance of text labels in the automatic text description tasks of a large number of video materials.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0031] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.
[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terms used herein in the specification of the application are for the purpose of describing specific embodiments only, and are not intended to limit the application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.
[0033] In the existing video description tasks, the number of generated labels ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a video material marking method and apparatus, a device and a medium. The method comprises the steps of obtaining a to-be-processed video material; performing key frame extraction processing on the video material based on a key frame extraction algorithm to obtain a key frame of the video material; the key frame extraction algorithm comprises a perceptual hash algorithm or a key frame extraction neural network algorithm; performing text labeling on the extracted key frame through a picture labeling algorithm to generate a description text corresponding to the key frame; performing word segmentation processing on the description text, and extracting subject words corresponding to the key frames; and performing duplicate removal processing on all the subject terms to obtain video visual content tags of the video materials. Compared with the traditional technology, the label generated by the scheme is not an element in a closed set, but is an open semantic label based on the visual content of the video, that is, the semantic text description of the visual content is generated, an open label system is finally output, and the effect of remarkably improving generalization performance is achieved.
Description
technical field [0001] The invention belongs to the technical field of video processing, and relates to a video material marking method, device, equipment and medium. Background technique [0002] With the continuous development of video processing technology, the methods for describing video data have become increasingly mature and widely used. The task of textual description of the video material is usually to translate the video content into the corresponding natural language, so as to realize the content tagging of the video material. The traditional video description methods on the market first pre-determine a variety of sentence templates, classify the words that make up the sentence according to their part of speech, obtain the descriptive words of the image through methods such as attribute learning and target recognition, and then combine the predicted words with the predefined sentences. The language model of template matching combines descriptive words, this meth...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06F16/783G06F16/75G06V20/40G06F40/30G06V10/82G06N3/04
CPCG06F16/7844G06F16/75G06F40/30G06N3/045
Inventor 吴一男丑晓慧
Owner 宁波深擎信息科技有限公司