Abstarct: Virtual reality is a new field in today's world. It has significant progress in several fields, including medicine, aerospace and marketing. Among the applications of virtual reality in the field of marketing, there is an interesting problem named virtual clothing try-on or virtual mirror. Virtual clothing try-on was proposed to solve the problems in the field of clothing and trying-on. When a customer goes shopping to buy a piece of clothes, he/she faces several problems such as the limitation in the number of clothes to try on, the lack of permission to try on bright clothes and the difficulty in putting clothes on for the customer. Virtual clothing try-on systems aim to simplify the task of trying-on virtually. The approaches in these systems can be divided into 2D and 3D mapping categories. Due to the limitations of the 2D-space, the 2D algorithms do not have a suitable output and perform poorly in transferring the clothes texture. On the other hand, the existing 3D algorithms are animated and far from reality. In this thesis, we proposed a new method for 3D virtual clothing which directly deals with the mapping of the 3D model of the selected clothes on the customer model. Among the challenges of 3D virtual dressing, we can mention preparing a 3D data without outliers, preparing a correct mesh, removing the head and neck area from the 3D model, execution speed, using the appropriate descxriptor and preserving the texture of the clothes. To provide a 3D model of the selected clothes, we put the clothes on a mannequin and capture its 3D point clothes while the mannequin is turning in front of the Kinect camera. In the proposed task, first the 3D model of clothes and customer is prepared by Kinect camera. Then, by obtaining the Laplace Beltrami descxriptor and using it, a model is acquired in which the customer sees the dress on his body. This is done by changing the size of the mannequin and 3D mapping. In this thesis, an algorithm is also introduced to estimate the size of the customer, which is useful in the final purchase of the customer. The output virtual dressing proposed in this thesis is more successful compared to the previous methods because, unlike the previous 2D methods, it is performing well in transferring the clothing texture and the output of the algorithm is not animated.