A visual special effect processing method and device, storage medium and terminal

Abstract

The present disclosure provides a visual special effect processing method and device, a storage medium and a terminal. The method includes: in response to receiving a special effect processing instruction, acquiring a multimedia image containing a portrait, and then processing the multimedia image using a segmentation network model to obtain a target motion area where the body part of the portrait is located, and, according to a preset grid Divide the target motion area into a plurality of grid areas, thereby, according to the displacement data of each grid area in the target motion area, determine the special effect action point, and then, the special effect action point in the multimedia image Click to display preset visual effects. The method of the present disclosure improves the matching degree of the effect point of the visual special effect and the body area of ​​the portrait, and improves the display effect of the visual special effect.

Description

technical field [0001] The present disclosure relates to visual special effect technology, and in particular, to a visual special effect processing method and device, a storage medium, and a terminal. Background technique [0002] With the popularization of smart terminals and the continuous development of visual effects technology, more and more applications focus on providing users with more personalized visual effects with better visual experience. [0003] In the processing of visual special effects for images containing portraits, it often involves identifying body regions (or clothing regions) of the portraits in the image, so as to display the effect of visual special effects on the body regions of the portraits. In the prior art, the identification of the body area of ​​a portrait is generally implemented by the pixel gradient method, that is, the body area and the background area are identified by comparing the pixel values ​​of adjacent pixels. [0004] However, t...

AI Technical Summary

Problems solved by technology

[0004] However, this recognition method has a good recognition effect on the junction of the body area and the background area, but it is difficult to accurately identify other body areas except the boundary.

As a result, the determined special effect points are concentrated at the junction of the body area and the background area, and the matching degree between the visual effect point and the body area of ​​the portrait is poor, which in turn leads to a poor display effect of the visual effect

Images