This Project was partially funded by the European Network of Excellence "PASCAL: Pattern Analysis, Statistical Modeling and Computational Learning", through the project "CARTER: Classification of visual Scenes using Affine invariant Regions and TExt Retrieval methods", and by the Swiss National Center of Competence in Research (NCCR) on Interactive Multimodal Information Management (IM)2, through the Swiss Federal Office for Education and Science (OFES), T. Tuytelaars is supported by the Fund for Scientific Research Flanders. We thank Luc Van Gool and Mihai Osian (Katholieke Universiteit Leuven, Belgium) for discussions.