Generative adversarial network-based high dynamic range inverse tone mapping method and system

Abstract

The present invention provides a generative adversarial network-based high dynamic range inverse tone mapping method and system. The method includes the following steps that: an original high dynamicrange video is read, and the original high dynamic range video is cut and converted into data sets which can be used for training and are corresponding to a standard dynamic range and a high dynamic range respectively; a generative adversarial network is established based on a convolutional neural network and a hopping type connection is established, and standard dynamic range images are convertedinto high dynamic range images, namely inverse tone mapping is performed; and the entire generative adversarial network is continuously optimized according to a set comprehensive objective function,and a finally obtained network can complete mapping from the standard dynamic range to the high dynamic range. With the method of the invention adopted, the problems such as non-linearity insufficiency and complicated parameter adjustment of an existing non-learning method can be solved; the one-dimensional characteristic and gradient characteristic of the high dynamic range images are considered,so that inverse dynamic mapping of the high dynamic range can be better realized.

Description

technical field [0001] The invention relates to a method in the technical field of image processing, in particular to a high dynamic range inverse tone mapping method and system based on a generative confrontation network. Background technique [0002] High Dynamic Range (HDR) is definitely a big trend in today's TV display and filming technology. High dynamic range is a very different technology from existing display technology, which is mainly reflected in the way it changes the picture. Thanks to the high dynamic range, the colors presented by the TV are more vivid, the blacks are deeper, and the objects in the picture are more clearly defined. At the same time, the color tone of the picture will also be expanded, from the BT.709 standard under the widely used standard dynamic range to the BT.2020 standard with high dynamic range. Brightness is the key to high dynamic range: most TVs on the market are around 400 nits, and some models hit 750 nits. But high dynamic rang...

AI Technical Summary

Problems solved by technology

[0003] Since none of the traditional methods are learning-based methods, their essence lies in segmented mapping, which has insufficient nonlinear performance and operability

Some learning-based methods that have emerged in recent years are mainly aimed at the expansion of the luminance gamut, and rarely consider the expansion of the color gamut, resulting in poor viewing experience

Images