Abstract: A video generation system and method is disclosed. The system includes a network, a camera device and a server. The camera device is for capturing an original video. The original video is transmitted by the network to the server. The server has a feature recognition unit, a medium object modification unit and a video synthesis unit. The feature recognition unit is for recognizing and positioning feature information of the original video. The medium object modification unit is for modifying a medium object based on the feature information to generate a modified medium object. The video synthesis unit is for synthesizing the original video and the modified medium object to generate a synthesized video based on the feature information.