Summary: ``Metal artifacts severely degrade image quality by generating streak artifacts in Xray computed tomography (CT) images. Metal artifact reduction (MAR) has long been an important issue because metal artifacts interfere with the acquisition of accurate contrast images, limiting the various applications of CT imaging. In this work, three recently developed CT MAR methods are introduced: normalized MAR, surgery-based MAR, and convolutional neural network-based MAR. Also, a MAR method for industrial cone beam CT is presented as an industrial application.''