Video acquisition method and device, video generation method and device

Abstract

The present invention relates to a video collection method and device, and a video generation method and device. The video collection method includes: acquiring video frames in various shooting directions required for constructing multi-view videos; mapping each video frame to a two-dimensional space, to obtain two-dimensional model data of each video frame in the two-dimensional space; mapping each video frame mapped to the two-dimensional space to a three-dimensional space according to a predetermined three-dimensional model, to obtain three-dimensional model data of each video frame in the three-dimensional space, and sending the video frame data and the model data of the video frame to a server to construct multi-view videos, wherein the model data includes the type of the three-dimensional model, the two-dimensional model data and the three-dimensional model data. According to the video collection method and device, and video generation method and device of the present invention, panoramic videos, any non-panoramic video or other continuous or discontinuous multi-view videos can be generated based on the acquired video frame data, the two-dimensional model data about each video frame, the three-dimensional model data, and the three-dimensional model, and distortion and deformation and the like can be reduced to a maximum extent, thereby improving the user experience.

Description

technical field [0001] The present invention relates to the field of video technology, in particular to a video collection method and device, and a video generation method and device. Background technique [0002] The collection and transmission of video information is an important direction for the development of information technology. Due to the limited field of view of a single camera, traditional video can only capture a certain layout of the scene, and cannot allow different users to watch scenes of interest from different angles at the same time. user's individual needs. However, panoramic video that has become popular in recent years overcomes the above-mentioned defects of traditional video. [0003] Panoramic video technology involves computer graphics, human-computer interaction technology, sensing technology, artificial intelligence and other fields. It uses computing methods to generate realistic three-dimensional visual and auditory sensations, and provides us...

AI Technical Summary

Problems solved by technology

[0004] At present, the video collector usually only provides the collected two-dimensional video data to the video generator, and the video generator creates a three-dimensional panoramic video according to a specific model (such as a sphere, a regular hexahedron, or a cone, etc.) based on the two-dimensional video data. However, the two-dimensional video data provided by the video acquisition party is usually only suitable for generating panoramic videos, and the video generator cannot conveniently and flexibly generate various discontinuous viewing angles based on the provided two-dimensional video data. Perspective video

And because the models used by different video generators are different, the two-dimensional video data collected by the video collector may not be compatible with the model used by the video generator, resulting in unsatisfactory video display effects (such as distortion, deformation, etc.) Wait)

Moreover, the existing technology can only generate panoramic videos based on existing 3D models such as spheres and regular hexahedrons. If you customize other types of models, it is more likely that the display will be reduced because the 2D video data cannot be adapted to the 3D model. Effect

Images