ABSTRACT
This paper addresses the construction of a short-vector (128D) image representation for large-scale image and particular object retrieval. In particular, the method of joint dimensionality reduction of multiple vocabularies is considered. We study a variety of vocabulary generation techniques: different k-means initializations, different descriptor transformations, different measurement regions for descriptor extraction. Our extensive evaluation shows that different combinations of vocabularies, each partitioning the descriptor space in a different yet complementary manner, results in a significant performance improvement, which exceeds the state-of-the-art.
- R. Arandjelovic and A. Zisserman. Three things everyone should know to improve object retrieval. In Proc. CVPR, pages 2911--2918, 2012. Google ScholarDigital Library
- R. Arandjelović and A. Zisserman. All about VLAD. In Proc. CVPR, 2013. Google ScholarDigital Library
- O. Chum and J. Matas. Unsupervised discovery of co-occurrence in sparse high dimensional data. In Proc. CVPR, 2010.Google ScholarCross Ref
- O. Chum, J. Philbin, J. Sivic, M. Isard, and A. Zisserman. Total recall: Automatic query expansion with a generative feature model for object retrieval. In Proc. ICCV, 2007.Google ScholarCross Ref
- P. Comon. Independent component analysis, a new concept? Signal processing, 36(3):287--314, 1994. Google ScholarDigital Library
- H. Jégou and O. Chum. Negative evidences and co-occurrences in image retrieval: the benefit of PCA and whitening. In Proc. ECCV, Firenze, Italy, Oct. 2012. Google ScholarDigital Library
- H. Jégou, M. Douze, and C. Schmid. On the burstiness of visual elements. In Proc. CVPR, 2009.Google ScholarCross Ref
- H. Jégou, M. Douze, and C. Schmid. Improving bag-of-features for large scale image search. IJCV, 87(3):316--336, 2010. Google ScholarDigital Library
- H. Jégou, M. Douze, and C. Schmid. Product quantization for nearest neighbor search. IEEE PAMI, 33(1):117--128, 2011. Google ScholarDigital Library
- H. Jégou, F. Perronnin, M. Douze, J. Sánchez, P. Pérez, and C. Schmid. Aggregating local image descriptors into compact codes. IEEE PAMI, 34(9):1704--1716, 2012. Google ScholarDigital Library
- H. Jégou, A. Zisserman, et al. Triangulation embedding and democratic aggregation for image search. In Proc. CVPR, 2014. Google ScholarDigital Library
- D. G. Lowe. Distinctive image features from scale-invariant keypoints. Proc. ICCV, 60(2):91--110, 2004. Google ScholarDigital Library
- J. Matas, O. Chum, M. Urban, and T. Pajdla. Robust wide baseline stereo from maximally stable extremal regions. In Proc. BMVC, volume 1, pages 384--393, 2002.Google ScholarCross Ref
- K. Mikolajczyk and C. Schmid. Scale & affine invariant interest point detectors. IJCV, 1(60):63--86, 2004. Google ScholarDigital Library
- A. Mikulik, M. Perd'och, O. Chum, and J. Matas. Learning vocabularies over a fine quantization. IJCV, pages 1--13, 2012. Google ScholarDigital Library
- D. Nister and H. Stewenius. Scalable recognition with a vocabulary tree. In Proc. CVPR, 2006. Google ScholarDigital Library
- A. Oliva and A. Torralba. Modeling the shape of the scene: A holistic representation of the spatial envelope. IJCV, 42(3):145--175, 2001. Google ScholarDigital Library
- M. Perdoch, O. Chum, and J. Matas. Efficient representation of local geometry for large scale object retrieval. In Proc. CVPR, 2009.Google ScholarCross Ref
- F. Perronnin, Y. Liu, J. Sanchez, and H. Poirier. Large-scale image retrieval with compressed fisher vectors. In Proc. CVPR, 2010.Google ScholarCross Ref
- J. Philbin, O. Chum, M. Isard, J. Sivic, and A. Zisserman. Object retrieval with large vocabularies and fast spatial matching. In Proc. CVPR, 2007.Google ScholarCross Ref
- J. Philbin, O. Chum, M. Isard, J. Sivic, and A. Zisserman. Lost in quantization: Improving particular object retrieval in largescale image databases. In Proc. CVPR, 2008.Google ScholarCross Ref
- F. Radenovic, H. Jegou, and O. Chum. Multiple Measurements and Joint Dimensionality Reduction for Large Scale Image Search with Short Vectors - Extended Version. ArXiv e-prints, Apr. 2015.Google Scholar
- J. Sivic and A. Zisserman. Video google: A text retrieval approach to object matching in videos. In Proc. ICCV, pages 1470--1477, 2003. Google ScholarDigital Library
- A. Torralba, R. Fergus, and Y. Weiss. Small codes and large image databases for recognition. In Proc. CVPR, pages 1--8. IEEE, 2008.Google ScholarCross Ref
- T. Tuytelaars and L. Van Gool. Wide baseline stereo matching based on local, affinely invariant regions. In Proc. BMVC, 2000.Google ScholarCross Ref
- Y. Weiss, A. Torralba, and R. Fergus. Spectral hashing. In Proc. NIPS, pages 1753--1760, 2009.Google ScholarDigital Library
Index Terms
- Multiple Measurements and Joint Dimensionality Reduction for Large Scale Image Search with Short Vectors
Recommendations
Comparative Study on Dimensionality Reduction in Large-Scale Image Retrieval
ISM '13: Proceedings of the 2013 IEEE International Symposium on MultimediaDimensionality reduction plays a significant role for the performance of large-scale image retrieval. In this paper, various dimensionality reduction methods are compared to validate their own performance in image retrieval. For this purpose, first, the ...
Multi-feature Image Retrieval by Nonlinear Dimensionality Reduction
ISCID '14: Proceedings of the 2014 Seventh International Symposium on Computational Intelligence and Design - Volume 02Multi-feature fusion is effective in raising the matching performance of image retrieval. However, the "Curse of Dimensionality" has to be solved. Traditional dimensionality reduction methods cannot reflect the high-order correlation among features and ...
Moments discriminant analysis for supervised dimensionality reduction
Most of the well-known supervised dimensionality reduction methods assume unimodal or Gaussian likelihoods, which may not be appropriate in the real life applications. In this manuscript, we introduce a novel supervised dimensionality reduction approach,...
Comments