A gesture recognition system includes an image capturing device, a memory unit and a processing unit. The image capturing device includes a zoom lens and captures an image frame with a focus length. The memory unit previously saves a lookup table of depths versus sharpness associated with at least one the focus length of the zoom lens. The processing unit is configured to calculate a current sharpness value of at least one object image in the image frame and to obtain a current depth of the object image according to the lookup table.