From images to scenes: Compressing an image cluster into a single scene model for place recognition.

Edward Johns, Guang-Zhong Yang

Browse the full ICCV paper archive.