Image warping method and computer program product thereof

Abstract

An image warping method and a computer program product thereof are provided. The image warping method comprises the following steps: defining a plurality of original feature points of an original image, wherein the original image corresponds to an original view angle; calculating a plurality of original pixel coordinates of the original feature points in the original image; defining a plurality of new feature points of the original image, wherein the new feature points respectively correspond to the original feature points; calculating a plurality of new pixel coordinates of the new feature points projected onto the original image; and approaching each of the original pixel coordinates of the original feature points to each of the new pixel coordinates of the corresponding new feature points in the original image so that the original image is warped into a new image, wherein the new image corresponds to a new view angle.

Description

technical field [0001] The present invention relates to a method for image deformation; more specifically, the present invention relates to a method for deforming an original image into a new image by approaching several original feature points of an original image to corresponding several new feature points image warping method. Background technique [0002] In response to the needs of modern people for 3D images, issues related to 3D images have gradually attracted people's attention, and in order to meet these needs, technologies related to 3D images are also increasingly refined. In recent years, stereoscopic image displays, such as three-dimensional television (Three-Dimensional Television; 3DTV), have been gradually popularized in the market, so that modern people can easily enjoy the visual enjoyment brought by stereoscopic images. However, due to technical considerations, stereoscopic image capture devices are not as popular as stereoscopic image display devices. T...

AI Technical Summary

Problems solved by technology

[0003] One of the main problems of the unpopularity of stereoscopic image capture equipment is that the technology for converting two-dimensional (Two-Dimensional Television; 2D) images into 3D images is not yet mature.

[0004] Unfortunately, the image depth information that the DIBR method relies on is not easy to obtain accurately

Generally speaking, image depth information can be obtained through manual processing or computer vision technology, but manual processing takes a lot of manpower and time, and computer vision technology also takes a long time to calculate

In addition, whether it is artificially or with computer vision technology, it is easy to estimate the image depth information accurately due to noise

On the other hand, the occlusion phenomenon between objects in an image will cause a hole after the image displacement of the new perspective, and the most criticized part of the DIBR method is that it must use adjacent pixels to fill the hole, so it is easy to produce virtual images. Edge and other issues

[0005] To sum up, the DIBR method is commonly used to convert most 2D images to 3D images at present, and the DIBR method is easily limited by the accuracy of image depth information, which makes it difficult to develop stereoscopic image capture technology

Images