A Light Field Image Compression Method Based on Macropixel Block Adaptive Prediction
An adaptive prediction, light field image technology, applied in image communication, digital video signal modification, electrical components, etc., can solve the problem of limited coding efficiency, increased computational complexity, and the failure of macro-pixel similarity in light field images to be effective. Utilize and other problems to achieve the effect of encoding efficiency, reducing redundant information, and accurate prediction
Active Publication Date: 2019-06-28
SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Among them, the image-based direct compression method uses the existing image or video encoder to directly compress the light field image, and introduces coding tools such as vector intra prediction and parallax compensation, which inevitably introduce block artifacts At the same time, the similarity relationship between the macropixels in the light field image has not been effectively utilized; while the indirect compression method based on the pseudo video sequence refers to extracting a series of sub-aperture images from the original light field data and generating pseudo Video sequence, using the existing video encoder to compress the video sequence, although this scheme improves the coding efficiency, but the computational complexity is also greatly increased
Both types of schemes have the following problems: due to the mismatch between the macro-pixel arrangement structure in the light field image and the block-based coding architecture of the encoder, existing coding tools cannot make full use of the correlation between macro-pixels, which limits the coding efficiency further improvement of
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0037] The present invention will be further described below with reference to the accompanying drawings and in combination with preferred embodiments.
[0038] Such as figure 1 As shown, the light field image compression method of the present invention is based on macropixel block adaptive prediction, comprising the following steps:
[0039] A1: Input an original light field image f(x, y), perform coordinate transformation and invalid pixel interpolation on the macro pixel blocks in the original light field image f(x, y), and obtain a regularized light field image L(x ,y); specifically:
[0040] The input original light field image f(x,y) is an orthogonally arranged microlens array image collected by a light field camera (such as Lytro Illum) after a series of pre-processing, such as Figure 2aAs shown, the image consists of a series of orthogonal macropixels. Each macro pixel in the image corresponds to a point in the actual scene, and it can simultaneously record the pos...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for compression of an optical field image based on adaptive prediction of macro-pixel blocks. The method comprises the following steps of inputting an original optical field image f(x, y) and processing the macro-pixel blocks in the original optical field image f(x, y) to obtain a regularized optical field image L(x, y); inputting the regularized optical field image L(x, y) into a video encoder, and selecting a predicted reference block Bi in a neighborhood according to the position and the size of a current coding block B0; obtaining a predicted value B'0 of the current coding block B0 through a linear weighted array prediction method and a single direct prediction method based on the current coding block B0 and the predicted reference block Bi respectively; and selecting the optimal coding prediction method to carry out compression on the optical field image. The method fully utilizes the arrangement characteristics of the macro-pixel blocks to realize accurate prediction of the current coding block, so that the redundant information in an airspace is effectively reduced, and thus the coding efficiency is effectively improved.
Description
technical field [0001] The invention relates to the fields of computer vision and digital image processing, in particular to a light field image compression method based on macro pixel block adaptive prediction. Background technique [0002] The hand-held light field camera can simultaneously record the position information and direction information of light under one exposure time, and the obtained light field data can be applied to many computer vision fields such as 3D stereoscopic display, saliency inspection, and depth estimation, showing extremely strong The application prospect has aroused widespread attention from people including academia. The light field image captured by the light field camera not only records the light angle and position information, but also has ultra-high resolution and special pixel distribution, which brings great challenges to data storage and transmission, and restricts the light field camera. development and application. Therefore, findi...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & AuthorityPatents(China)
IPC IPC(8): H04N19/103H04N19/147H04N19/182
CPCH04N19/103H04N19/147H04N19/182
Inventor金欣韩海旭戴琼海
OwnerSHENZHEN GRADUATE SCHOOL TSINGHUA UNIV