文章目录
维基百科
中文版
基准真相：(ground truth) 是一个相对概念；
它是指相对于新的测量方式得到的测量值，作为基准的，由已有的、可靠的测量方式得到的测量值（即经验证据）。人们往往会利用基准真相，对新的测量方式进行校准，以降低新测量方式的误差和提高新测量方式的准确性。
机器学习领域借用了这一概念。使用训练所得模型对样本进行推理的过程，可以当做是一种广义上的测量行为。因此，在有监督学习中，ground truth 通常指代样本集中的标签。
Statistics and machine learning
“Ground truth” may be seen as a conceptual term relative to the knowledge of the truth concerning a specific question. It is the ideal expected result.[2] This is used in statistical models to prove or disprove research hypotheses. The term “ground truthing” refers to the process of gathering the proper objective (provable) data for this test. Compare with gold standard. For example, suppose we are testing a stereo vision system to see how well it can estimate 3D positions. The “ground truth” might be the positions given by a laser rangefinder which is known to be much more accurate than the camera system.
Bayesian spam filtering is a common example of supervised learning. In this system, the algorithm is manually taught the differences between spam and non-spam. This depends on the ground truth of the messages used to train the algorithm – inaccuracies in the ground truth will correlate to inaccuracies in the resulting spam/non-spam verdicts.
I Can See Clearly Now : Image Restoration via De-Raining
我正在阅读论文"I Can See Clearly Now : Image Restoration via De-Raining"，作者：Horia Porav, Tom Bruls and Paul Newman;
论文中用这个词语指代一个基准的图片，和优化去雨滴之后的图片来做对比，发现本文提出的方法切实可行。