Efficient image signatures and similarities using tensor products of local descriptors

David Picard (1) , Philippe-Henri Gosselin (1)

1 MIDI - Multimedia Indexation and Data Integration

In this paper, we introduce a novel image signature effective in both image retrieval and image classification. Our approach is based on the aggregation of tensor products of discriminant local features, named VLAT (vector of locally aggregated tensors). We also introduce techniques for the packing and the fast comparison of VLATs. We present connections between VLAT and methods like kernel on bags and Fisher vectors. Finally, we show the ability of our method to be effective for two different retrieval problems, thanks to experiments carried out on similarity search and classification datasets.

Domains: Machine Learning [stat.ML]; Multimedia [cs.MM]

David Picard, Philippe-Henri Gosselin. Efficient image signatures and similarities using tensor products of local descriptors. Computer Vision and Image Understanding, 2013, 117 (6), pp.680-687. ⟨10.1016/j.cviu.2013.02.004⟩. ⟨hal-00799074⟩