Generation method and system for video frame depth chart

Abstract

The invention provides a generation method and system for video frequency frame depth graph. The method comprises: selecting critical frame in continuous video frequency frame of vide frequency sequence, and acquiring depth graph of critical frame; also following steps, extracting characteristic point of continuous frame of video frequency sequence; determining matching relation of each characteristic point of continuous frame; computing movement distance between characteristic points matching at continuous frame; then determining the corresponding characteristic point depth value at non-critical frame; performing depth assign for the non-critical frame, further acquiring depth graph of every video frequency frame of the continuous frame. The inventive method and system acquires relatively accurate depth graph of video frequency frame without need of manpower, largely shortens production period from planar video frequency to stereo video frequency, and reduces production cost and simultaneously promotes efficiency.

Description

technical field [0001] The invention relates to the technical field of computer multimedia, in particular to a method and system for generating a depth map of a video frame in a video sequence when a plane video is converted into a stereoscopic video. Background technique [0002] Stereoscopic video, compared with the widely used flat video, can provide users with the depth information of the scene in the video, so that users can see the scene almost completely consistent with the real world through the video, resulting in a huge sense of reality and presence. Because of this advantage, stereoscopic video is being promoted and favored more and more in many industries all over the world. [0003] With the development of stereoscopic video technology, the source production of stereoscopic video has received great attention from people. The technical means of transforming planar video into stereoscopic video has become an important method to obtain stereoscopic video source be...

AI Technical Summary

Problems solved by technology

[0007] However, for the fully automatic depth map generation method, since the texture, motion and other information are not necessarily related to the depth value, this method cannot obtain an accurate depth map

[0008] For the semi-automatic depth map generation method, since the frame rate of the general video is above 20 frames per second, if human-computer interaction is performed on each frame of a video to obtain the depth map, the workload will be extremely heavy. It will prolong the production cycle of stereoscopic video and increase the production cost

Images