Image-based image rendering method and system using multi-camera and depth camera arrays

Abstract

A method and system for capturing imagery and depth for image-based rendering. Capture with a multi-camera structure by using a combination of an image camera and a depth camera. Rendering utilizes scene geometry derived from image and depth data.

Description

technical field [0001] The present invention relates to the field of image technology, and in particular, to an image-based image rendering method and system using a multi-camera and depth camera array. Background technique [0002] Images and videos can be created using a variety of techniques. For image-based rendering, where images are created based on other images, excluding the number of images that can be sampled, there can be problems determining certain contextual pieces of information. In order to improve the quality or accuracy of two-dimensional (2D) or three-dimensional (3D) images or videos from a human visual perspective, individual images need to be rendered by understanding context such as the depth of objects or scene geometry in the image. However, this is nearly impossible to accomplish unless additional information is provided. [0003] In non-image-based computer-generated renderings (eg, rasterization, ray tracing, etc.) that use geometric models, whe...

AI Technical Summary

Problems solved by technology

However, just like when an image camera is out of focus, when the image is not correctly reconstructed from various inputs (in this case from other images), that reconstructed image may appear out of focus or out of focus

Alternatively, the rendered image may have artifacts or other image errors

Images