图像质量评价相关论文阅读笔记（二）：Controllable List-wise Ranking for Universal No-reference Image Quality Assessment

Controllable List-wise Ranking for Universal No-reference Image Quality Assessment

Authors: Fu-Zhao Ou, Yuan-Gen Wang, Jin Li, Guopu Zhu, Sam Kwong

Abstract: No-reference image quality assessment (NR-IQA) has received increasing attention in the IQA community since reference image is not always available. Real-world images generally suffer from various types of distortion. Unfortunately, existing NR-IQA methods do not work with all types of distortion. It is a challenging task to develop universal NR-IQA that has the ability of evaluating all types of distorted images. In this paper, we propose a universal NR-IQA method based on controllable list-wise ranking (CLRIQA). First, to extend the authentically distorted image dataset, we present an imaging-heuristic approach, in which the over-underexposure is formulated as an inverse of Weber-Fechner law, and fusion strategy and probabilistic compression are adopted, to generate the degraded real-world images. These degraded images are label-free yet associated with quality ranking information. We then design a controllable list-wise ranking function by limiting rank range and introducing an adaptive margin to tune rank interval. Finally, the extended dataset and controllable list-wise ranking function are used to pre-train a CNN. Moreover, in order to obtain an accurate prediction model, we take advantage of the original dataset to further fine-tune the pre-trained network. Experiments evaluated on four benchmark datasets (i.e. LIVE, CSIQ, TID2013, and LIVE-C) show that the proposed CLRIQA improves the state of the art by over 8% in terms of overall performance. The code and model are publicly available at https://github.com/GZHU-Image-Lab/CLRIQA.

1.论文说是开源，但是github中并没有源代码。GitHub已经开源。

2.数据增强

    论文的原理和RankIQA类似，都是通过人工的方式增加数据集。除了常规的噪声、模糊等处理，还在上述处理过的数据上进行了过曝和曝光不足的操作，曝光的强度按照Weber-Fechner定律为10%。数据增加倍数从15（降质量操作）*5（操作的参数）倍增加到15*5*3（三种曝光操作）倍。

3.loss设计

   论文采用了liet_wise而不是pair_wise的rank方式。论文做了单独的比对试验证明CLR的方式比PWR的方式，在LIVE_C上SROCC提升一个点。五组对比试验，证明了数据扩增、Loss中添加margin和upper-lower bound对图像质量的学习更有效果。