Bundle min-hashing for logo recognition

We present a scalable logo recognition technique based on feature bundling. Individual local features are aggregated with features from their spatial neighborhood into bundles. These bundles carry more information about the image con-tent than single visual words. The recognition of logos in novel images is then performed by querying a database of ref-erence images. We further propose a novel WGC-constrained ransac and a technique that boosts recall for object re-trieval by synthesizing images from original query or refer-ence images. We demonstrate the benefits of these tech-niques for both small object retrieval and logo recognition. Our logo recognition system clearly outperforms the current state-of-the-art with a recall of 83 % at a precision of 99%