Network advertisement promotion method based on video content
A technology for video content and online advertising, which is applied in the field of online advertising promotion based on video content, can solve the problems of large cost investment, viewer dislike, and expensive manual operation, so as to improve the utilization rate of funds, reduce manual operations, and reduce The effect of promotion costs
Pending Publication Date: 2020-08-28
LINGTONG EXHIBITION SYST +1
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
This type of advertisement can generally be divided into two types, one is not related to the content of the video, which is likely to arouse the disgust of the viewer, but will have a counterproductive effect; the other is related to the content of the video, but requires a large number of staff to pre-set, Requires a lot of manual work
Not only the cost is large, but also affects the actual publicity effect of the advertisement
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 1
[0017] Embodiment 1: This embodiment comprehensively applies the theory and technology of residual deep learning, regional convolutional neural network learning, recurrent adversarial neural network learning, and text string similarity matching to develop and form a new type of online advertising promotion based on video content method, the details are as follows:
[0018] (1) Sampling of video sequences and matching method of video content and advertising model: First, collect as many videos containing subtitles as possible, construct a video sequence sampling model, and extract key frames of video sequences; then according to the sampled pictures containing subtitles Sequence, using residual deep learning to build a matching model for video content and advertisements.
[0019] (2) Embed the selected advertisement into the video sequence seamlessly: first, detect the target in the video sequence according to the regional convolutional neural network, extract the target featur...
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 network advertisement promotion method based on video content. The method comprises the following steps: (1) sampling a video sequence; (2) constructing a video content andadvertisement matching model; (3) seamlessly embedding the selected advertisement into a video sequence: detecting a target in the video sequence according to a regional convolutional neural network,extracting target features in the video sequence, and selecting an appropriate advertisement to be matched with the selected advertisement in combination with subtitles in the video sequence; and constructing a constraint adversarial neural network, and converting a specific target in the sequence into product advertisement information by utilizing the constraint adversarial neural network. According to the method, theories and technologies of residual deep learning, regional convolutional neural network learning, cyclic adversarial neural network learning and text string similarity matching are comprehensively applied; the method changes the promotion mode of the current product advertisement, flexibly uses the network to promote the product, improves the popularity of the product, obviously reduces the promotion cost of the product, and improves the utilization rate of funds.
Description
technical field [0001] The invention belongs to the technical field of network advertisement information processing, and in particular relates to a novel network advertisement promotion method based on video content. Background technique [0002] With the advent of the intelligent era, intelligence has had an important and far-reaching impact on the development of various industries in China. For China's advertising and exhibition industry, the application of intelligence not only brings convenience, but also creates new business opportunities. [0003] With the rapid development of communication computer network technology in recent years, the rapid popularization of smart portable devices, each mobile phone customer is a smart terminal, more and more users watch online video programs through smart devices (mobile phones, tablets) to increase their understanding of the outside world , Improve user entertainment and science. However, while the video is being watched, there...
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): H04N21/81H04N21/488H04N21/466G06Q30/02G06N3/04G06K9/62
CPCH04N21/812H04N21/4666H04N21/4884G06Q30/0271G06N3/045G06F18/22G06F18/214
Inventor 周军相入喜蔡飞
Owner LINGTONG EXHIBITION SYST