Audience evaluation data-driven silent product video creation auxiliary method and device

Abstract

The invention discloses an audience evaluation data-driven silent product video creation auxiliary method, which comprises the following steps that: label consistency screening is performed on a video initial sample set to obtain a product video sample set, the product video sample set is divided into an appearance type product video sample set and a function type product video sample set based on product categories, a high-quality case library is constructed, and an evaluation prediction model is constructed through an XGBoost method; an evaluation and prediction model interpreter is constructed through an SHAP algorithm, high-quality cases matched with materials, video attributes and product types provided by a user are screened out from a high-quality case library, high-quality characteristics of the high-quality cases are obtained through the prediction model interpreter to assist the user in completing video creation, and evaluation results of video fragments are predicted through an evaluation and prediction model; and a model interpreter is used to obtain the feature that the SHAP value of the video fragment is negative and the target value domain of feature optimization, and a user is assisted to adjust the feature value of the feature so as to complete optimization of the video fragment. According to the method, high-quality reference cases can be provided, and evaluation and optimization prompting can be carried out on video slices made by a user.

Description

technical field [0001] The invention belongs to the technical field of computer vision and computer-aided design, and in particular relates to an auxiliary method and device for video creation of silent products driven by audience evaluation data. Background technique [0002] With the upgrading of Internet consumption, the way for merchants to compete for the market is no longer just "price war", but to improve the consumption experience. Therefore, in recent years, intuitive and rich product videos have gradually become the most popular product display method on e-commerce platforms. The product video, also known as the main image video, is located on the first screen of the details page. Its main function is to help consumers obtain rich and in-depth product information efficiently and comfortably, thereby promoting purchase decisions. Compared with the graphic introduction of the previous e-commerce platform, the information carrying capacity of the product video is lar...

AI Technical Summary

Problems solved by technology

In addition, the previous method mainly completed the function of predicting the evaluation results, and could not provide creators with insufficient feature analysis. Therefore, it was difficult to help novices to optimize the video, and they could not solve the practical problems of novices in the production process. Content screening for the platform

Images