A method and a system for generating a video of a 3D object are provided. The method comprises: generating a visual reference frame; capturing an image of a surrounding area around the 3D object; in response to detecting within the image at least a portion of the 3D object: generating a bounding box encompassing the at least the portion of the 3D object; determining an overlap parameter between the bounding box and the visual reference frame; comparing the overlap parameter to a plurality of thresholds; in response to the overlap parameter being above a given one of the plurality of thresholds, generating a respective one of a plurality of predetermined indications for adjusting a position of the camera to capture the portion of the 3D object within the visual reference frame, thereby generating an adjusted image thereof used for composing the video of the 3D object.