A Depth Image Super-Resolution Method

Abstract

The invention discloses a depth map super-resolution method. The method comprises the following steps: carrying out training of a low-resolution dictionary AL and a high-resolution dictionary AH; extracting a low-resolution image block set (as shown in the specification) and a corresponding low-resolution characteristic set (as shown in the specification) from a low-resolution depth map D'L to be processed; for each low-resolution characteristic (as shown in the specification), inputting a noise-free high-resolution guiding color map CH; calculating the weight constraint W satisfied by each high-resolution characteristic by utilizing the CH; by combining the low-resolution dictionary AL, the high-resolution dictionary AH and the weight constraint W, solving by utilizing a 0 norm to obtain a reconstruction coefficient alpha; reconstructing corresponding high-resolution characteristics (as shown in the specification) by utilizing the reconstruction coefficient alpha and the high-resolution dictionary AH; adding all the corresponding high-resolution image blocks (as shown in the specification) to all the reconstructed high high-resolution characteristics (as shown in the specification) to obtain all the corresponding high-resolution image blocks (as shown in the specification); and fusing all the high-resolution image blocks to form a high-resolution depth map D'H. According to the method, a guiding image interpolation method and a sparse representation method are combined, so that the processing speed is high and the obtained high-resolution depth map is high in quality.

Description

technical field [0001] The invention relates to the fields of computer vision and image processing, in particular to a depth map super-resolution method. Background technique [0002] Depth maps, as a representation of the location of objects in real 3D space, play an increasingly important role in areas such as object recognition, stereo medicine, 3D reconstruction, and stereo video. Although there are currently many depth acquisition methods that can obtain depth information from natural scenes to form a depth map, the resolution of the obtained depth map is equivalent to that of the color map taken in the same scene. Noise, which is mainly due to the small sensing chip size of the acquisition devices used by these depth acquisition methods. These shortcomings greatly limit the development of practical applications based on depth information. [0003] Depth map super-resolution is an important means to solve the above problems. Depth image super-resolution belongs to th...

AI Technical Summary

Problems solved by technology

Such methods do not incorporate guided imagery and do not deal well with noise

Images