TM-Net: A Neural Net Architecture for Tone Mapping.

TM-Net: A Neural Net Architecture for Tone Mapping.
复制标题

DOI:
10.3390/jimaging8120325
复制
发表时间:
2022-12-12
期刊:
影响因子:
3.2
通讯作者:
--
中科院分区:
其他
文献类型:
--
作者:

文献摘要

参考文献

相似文献

色调映射函数应用于图像,以压缩图像的动态范围,使图像细节更加明显,最重要的是,产生令人愉悦的再现效果。对比度有限直方图均衡 (CLHE) 是最简单且部署最广泛的色调映射算法之一。 CLHE 的工作原理是迭代细化输入直方图(以满足某些条件)直至收敛,然后结果的累积直方图用于定义用于增强图像的色调图。本文做出了三点贡献。首先,我们证明 CLHE 可以精确地表述为深度色调映射神经网络(我们称之为 TM-Net)。 TM-Net 具有与 CLHE 中的细化一样多的层(即 60 多个层,因为 CLHE 最多需要 60 次细化才能收敛)。其次,我们证明我们可以训练一个固定的 2 层 TM-Net 来计算 CLHE,从而使 CLHE 的计算速度提高 30 倍。第三,我们采用了更复杂的色调映射器(使用二次规划),并表明它也可以使用定制的训练有素的 2 层 TM-Net 来实现,而不会损失视觉准确性。在包含 40,000 多张图像的大型语料库上进行的实验验证了我们的方法。
Tone mapping functions are applied to images to compress the dynamic range of an image, to make image details more conspicuous, and most importantly, to produce a pleasing reproduction. Contrast Limited Histogram Equalization (CLHE) is one of the simplest and most widely deployed tone mapping algorithms. CLHE works by iteratively refining an input histogram (to meet certain conditions) until convergence, then the cumulative histogram of the result is used to define the tone map that is used to enhance the image. This paper makes three contributions. First, we show that CLHE can be exactly formulated as a deep tone mapping neural network (which we call the TM-Net). The TM-Net has as many layers as there are refinements in CLHE (i.e., 60+ layers since CLHE can take up to 60 refinements to converge). Second, we show that we can train a fixed 2-layer TM-Net to compute CLHE, thereby making CLHE up to 30× faster to compute. Thirdly, we take a more complex tone-mapper (that uses quadratic programming) and show that it too can also be implemented — without loss of visual accuracy—using a bespoke trained 2-layer TM-Net. Experiments on a large corpus of 40,000+ images validate our methods.
DOI: 10.1109/tip.2012.2226047
发表时间: 2013-03-01
影响因子: 10.6
作者:
Huang, Shih-Chia;Cheng, Fan-Chieh;Chiu, Yi-Sheng
通讯作者: Chiu, Yi-Sheng
DOI: 10.1016/0141-9382(89)90009-7
发表时间: 1989-07-01
期刊: DISPLAYS
影响因子: 4.3
作者:
MEYER, GW
通讯作者: MEYER, GW
DOI: 10.1109/30.580378
发表时间: 1997-02-01
影响因子: 4.3
作者:
Kim, YT
通讯作者: Kim, YT
DOI: 10.1073/pnas.83.10.3078
发表时间: 1986-05-01
影响因子: 11.1
作者:
LAND, EH
通讯作者: LAND, EH
DOI: 10.1109/tip.2009.2021548
发表时间: 2009-09-01
影响因子: 10.6
作者:
Arici, Tarik;Dikbas, Salih;Altunbasak, Yucel
通讯作者: Altunbasak, Yucel