Depth/disparity map post-processing method and apparatus

Abstract

A depth / disparity map post-processing method and apparatus adopt the idea of combining edge information with segment information during irregular edge region detection. The method comprises: performing super-pixel segmentation of a color image when the color image is divided into image blocks; dividing a gray range into a preset number of intervals, and obtaining a histogram in the interval where all the pixels fall through statistics for each super pixel; and determining, in the current super pixel, whether a ratio of the number of pixels in the interval with the maximum interval distribution value to the number of all pixels in the current super pixel is less than a first threshold value, and if so, performing further segmentation of the current super pixel by using the color block dividing method. The invention improves the accuracy of color image segmentation while ensuring the image processing speed, thereby enhancing the accuracy of the irregular edge region detection.

Description

technical field [0001] The present application relates to the technical field of three-dimensional image processing, and in particular to a method and device for post-processing a depth map / disparity map. Background technique [0002] With the development of technology and the continuous improvement of people's needs, the acquisition of information from the outside world has become increasingly important. From the earliest black-and-white photos to color photos, to videos that can record time information, record and show the world people live in. The means are constantly improving, and the 3D technology that has begun to appear in recent years has greatly improved the way humans perceive the world. Applications such as 3D movies, naked-eye TV, virtual reality, and augmented reality have greatly enriched people's lives. At the same time, it also makes some scientific research more convenient. The key difference between these applications and previous applications is the addi...

AI Technical Summary

Problems solved by technology

[0005] Based on the above, it can be concluded that depth information, as a key technology in many cutting-edge fields and new applications, has received extensive attention and there are a large number of methods for obtaining depth information. However, due to technical limitations, the quality of depth maps will There are many problems. Some methods have begun to study the post-processing of the depth map. However, the processed image still has black holes and irregular object edges, which will seriously affect subsequent applications.

Images