Conventional reverse engineering uses 3D scanning or image processing technologies to extract the shape information of a given object in terms of point clouds before creating its virtual and real models. It requires complex and user-intensive geometric modeling processes such as noise removal and surface reconstruction. In order to avoid all these complex processes, this paper presents a scanning-free reverse engineering approach defined as a human-cognition-based reverse engineering. The approach consists of four spaces. The first space is defined as a human cognition space wherein the information of a given object is represented using some predefined linguistic expressions. The second space is defined as a point cloud creation space wherein the predefined linguistic expressions are translated to a point cloud model. The third space is defined as a virtual model creation space wherein the point cloud model is converted into a virtual model using a commercially available CAD system. The last space is defined as a real model creation space wherein the virtual model is converted into a real model using 3D printing technology. The efficacy of the proposed approach is demonstrated by performing a case study. The findings will help those who want to simplify the current practices regarding reverse engineering.