Google Patent | Conversion of 2D image to 3D video
Publication Number: 10154242
Publication Date: 2018-12-11
A two-dimensional input image to be used in a creation of a three-dimensional video may be received and depth values for pixels in the image may be determined. A depth map may be generated based on the depth values for the pixels and pixel shift values for the pixels may be calculated based on the depth map and a view disparity value. A modified image corresponding to a particular frame of the three-dimensional video may be generated based on the input image and the pixel shift values. An additional modified image corresponding to a next frame of the three-dimensional video may be generated based on the modified image and the pixel shift values used to generate the modified image where the modified image in combination with the input image and the additional modified image are a sequence of frames in the three-dimensional video.
Advances in display technology have caused display devices capable of conveying a stereoscopic perception of three-dimensional (3D) depth to a view to become increasingly common. These 3D displays may be found in high-definition (HD) television sets, gaming devices, and other computing devices. The growing number of 3D displays has led to a need for additional 3D visual content (e.g., images, video). Conventionally, creating 3D content has been a difficult and time consuming process. This typically includes a lengthy, highly technical and expensive manual process. Conversion of two-dimensional (2D) images to 3D video has been performed; however, conventional conversion techniques use at least two images of the same subject, from different viewpoints, to create each frame in the resulting 3D video. In addition, conventional techniques are limited to converting specific types of images and may not be used for general 2D to 3D conversion tasks.
The following is a simplified summary of the disclosure in order to provide a basic understanding of some aspects of the disclosure. This summary is not an extensive overview of the disclosure. It is intended to neither identify key or critical elements of the disclosure, nor delineate any scope of the particular implementations of the disclosure or any scope of the claims. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that is presented later.
In one implementation, a 3D video generator receives a two-dimensional input image to be converted to a first three-dimensional video. The video generator determines a depth value for each of a plurality of pixels in the original image and generates a depth map based on the depth values. The video generator can generate a modified image based on the input image, the depth map, and a view disparity. The original input image in combination with the modified image can form a sequence of frames in the three-dimensional video.
In one implementation, a depth value is determined for each of a plurality of pixels in an image by identifying a color value for each pixel, and identifying an estimated depth value that is associated with the color value. The color values may be defined, for example, using the Y-Cb-Cr color scheme or some other color scheme. The depth values for each pixel can be combined together to form a depth map. The modified image can be created by filtering the depth map, calculating a pixel shift for pixels in a modified image based on the depth values and the view disparity, and removing or filling in voids in the resulting modified image. At least one of a smoothing filter or a sharpening filter can be used to refine the depth map. Calculating a pixel shift for each of a plurality of pixels can be done by applying depth values and the view disparity to a view rendering algorithm. The output of the view rendering algorithm may define a pixel shift value for each pixel in the input image. In the modified image, pixels may be removed that are hidden from view due to the change in view disparity or added if they are visible.