Vision based campus guide system on intelligent mobile phone

In this paper, we propose a vision based campus guide system to provide visitors guide information on the campus. Images of buildings are firstly represented by visual words. Then, to further reduce the recognition time, visual words of the same building are merged to a distinct codebook. During visiting, visitors capture images of an interested building by their mobile phones for recognition. After recognizing the building, tour guide information will be sent to the visitors via internet. Experimental results show that our method can efficiently recognize the target buildings under various viewpoints.