**5. Conclusions**

In this study, a human-height estimation method using color and depth information was proposed. The human body region was extracted through the pre-trained mask R-CNN to color video. The human body region extraction from depth video was also proposed by comparing with the background depth image. Human height was estimated from depth information by converting two points of head-top and foot-bottom into two 3D real-world coordinates and by measuring the Euclidean distance between two 3D coordinates. Human height was accurately estimated even if the person is not in front or a walking state. In the experiment results, the errors of the human-height estimation by the proposed method with the standing state were 0.7% and 2.2% when the human body region was extracted by mask-R CNN and by the background depth image, respectively. The proposed method significantly improves the human-height estimation by combining color and depth information. The proposed method can be applied to estimate not only the body height, but also the height of other object types such as animals. The proposed method can also be applied to gesture recognition and body posture estimation which require the types and the 3D information of objects.

**Author Contributions:** Conceptualization, D.-s.L., J.-s.K., S.C.J. and S.-k.K.; software, D.-s.L. and J.-s.K.; writing—original draft preparation, D.-s.L., J.-s.K., S.C.J. and S.-k.K.; supervision, S.-k.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the BB21+ Project in 2020 and was supported by Dong-eui University Foundation Grant (2020).

**Conflicts of Interest:** The authors declare no conflict of interest.
