Bag-of-Bags of Words model over irregular graph partitions for image retrieval

Yi Ren (1) , Aurélie Bugeau (1) , Jenny Benois-Pineau (1)

1 LaBRI - Laboratoire Bordelais de Recherche en Informatique

The paper presents a novel approach, named bag-of-bags of words (BBoW), to address the problem of Content-Based Image Retrieval (CBIR) from image databases. The proposed bag-of-bags of words model extends the classical bag-of-words (BoW) model. An image is represented as a graph of local features on a regular grid. Then irregular partitions of images are built using different graph cutting methods. Each graph is then represented by its own signature. Compared to existing methods for image retrieval, such as Spatial Pyramid Matching (SPM), the BBoW model does not assume that similar parts of a scene always appear at the same location in images of the same category. The extension of the proposed model to pyramid gives rise to a method we name irregular pyramid matching. The experiments demonstrate the strength of our method for image retrieval when the partitions are stable across an image category. The experimental results for Caltech101 benchmark show that our method achieves comparative results as SPM, and is globally more stable.

Keywords: Normalized Cuts; Content-based image retrieval; Clustering; Graph partitioning; Segmentation; Bag of words; Graph Cuts; Normalized Cuts.

Domains: Computer Vision and Pattern Recognition [cs.CV]; Image Processing [eess.IV]; Information Retrieval [cs.IR]

Yi Ren, Aurélie Bugeau, Jenny Benois-Pineau. Bag-of-Bags of Words model over irregular graph partitions for image retrieval. 2013. ⟨hal-00976939⟩