Generation method and device of depth map through three-dimensional conversion to planar video

Abstract

The invention provides generation method and device of a depth map through three-dimensional conversion to planar video, wherein the method comprises the following steps of: judging whether an image is effective data; carrying out grayscale conversion to the effective data to acquire a gray value if the image is the effective data; calculating an initial depth value of the image by applying a gray clue according to the gray value; calculating the depth value of an image parallax point according to a zero parallax point template; calculating a final depth map of the image according to the initial depth value and the depth value of the zero parallax point; and carrying out Gaussian filtering to the final depth map to acquire a smooth depth map of the image. The method overcomes the defects of slow processing speed, poor real-time and bad compatibility of the prior art and greatly increases the generation speed of the depth map, and can meet requirements on real-time display of a plurality of display equipment because of small volume of the device, low cost and strong compatibility, and a large number of planar film sources is converted into three-dimension film sources to solve the problem of being lack of three-dimension film sources.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a method and device for generating a depth map for stereo conversion of planar video based on programmable devices. Background technique [0002] Stereoscopic video can bring viewers a realistic viewing experience. By making viewers see images with parallax in both eyes, a stereoscopic effect is produced, allowing viewers to have a wonderful feeling of being on the scene. In recent years, stereoscopic video has been familiar to and loved by more and more viewers. Some well-made movies have achieved great success with the help of related technologies of stereoscopic video, creating beautiful scenes that are unforgettable for the audience. It can be said that stereoscopic video, with its unique viewing experience, allows viewers to experience the feeling of being in the video scene, and is unanimously recognized by the majority of viewers, and has become the main direction of video ...

AI Technical Summary

Problems solved by technology

[0007] The present invention aims at the disadvantages of slow processing speed, poor real-time performance, and poor compatibility in the prior art and methods, and proposes a method and device for generating a stereoscopic conversion depth map of a plane video. The method is based on a single plane image. The automatic depth map generation method does not require manual participation in the processing process, and the device can meet various video sources as input, and provide real-time depth information for stereoscopic display devices in various modes, with strong compatibility

Images