Exploiting contextual information for image re-ranking and rank aggregation

Daniel Carlos Guimarães Pedronette1, Ricardo da S. Torres1
1RECOD Lab., Institute of Computing (IC), University of Campinas (UNICAMP), Campinas, Brazil

Tóm tắt

Content-based image retrieval (CBIR) systems aim to retrieve the most similar images in a collection, given a query image. Since users are interested in the returned images placed at the first positions of ranked lists (which usually are the most relevant ones), the effectiveness of these systems is very dependent on the accuracy of ranking approaches. This paper presents a novel re-ranking algorithm aiming to exploit contextual information for improving the effectiveness of rankings computed by CBIR systems. In our approach, ranked lists and distance scores are used to create context images, later used for retrieving contextual information. We also show that our re-ranking method can be applied to other tasks, such as (a) combining ranked lists obtained using different image descriptors (rank aggregation) and (b) combining post-processing methods. Conducted experiments involving shape, color, and texture descriptors and comparisons with other post-processing methods demonstrate the effectiveness of our method.

Tài liệu tham khảo

Abowd GD, Dey AK, Brown PJ, Davies N, Smith M, Steggles P (1999) Towards a better understanding of context and context-awareness. In: Proceedings of the 1st international symposium on handheld and ubiquitous Computing, HUC’99, pp 304–307 Arica N, Vural FTY (2003) Bas: a perceptual shape descriptor based on the beam angle statistics. Pattern Recogn Lett 24(9–10):1627–1639 Bai X, Wang B, Wang X, Liu W, Tu Z (2010) Co-transduction for shape retrieval. ECCV 3:328–341 Belongie S, Malik J, Puzicha J (2002) Shape matching and object recognition using shape contexts. PAMI 24(4):509–522 Brodatz P (1966) Textures: a photographic album for artists and designers. Dover, USA Coppersmith D, Fleischer LK, Rurda A (2010) Ordering by weighted number of wins gives a good ranking for weighted tournaments. ACM Trans Algorithms 6:55:1–55:13 Croft WB (2002) Combining approaches to information retrieval. In: Croft WB (ed) Advances in information retrieval. The information retrieval, vol 7. Springer, USA, pp 1–36 da S Torres R, Falcão AX (2006) Content-based image retrieval: theory and applications. Revista de Informática Teórica e Aplicada 13(2):161–185 da Torres RS, Falcão AX (2007) Contour salience descriptors for effective image retrieval and analysis. Image Vis Comput 25(1):3–13 Dai HJ, Lai PT, Tsai RTH, Hsu WL (2010) Global ranking via data fusion. In: Proceedings of the 23rd international conference on computational linguistics: posters, COLING’10, pp 223–231 Datta R, Joshi D, Li J, Wang JZ (2008) Image retrieval: ideas, influences, and trends of the new age. ACM Comput Surv 40:5:1–5:60 Diaz F (2005) Regularizing ad hoc retrieval scores. In: CIKM’05, pp 672–679 Dwork C, Kumar R, Naor M, Sivakumar D (2001) Rank aggregation methods for the web. In: Proceedings of the 10th international conference on World Wide Web. ACM, New York, WWW’01, pp 613–622. doi:10.1145/371920.372165 Fagin R, Kumar R, Mahdian M, Sivakumar D, Vee E (2004) Comparing and aggregating rankings with ties. In: Proceedings of the twenty-third ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems, PODS’04, pp 47–58 Faria FF, Veloso A, Almeida HM, Valle E, da S Torres R, Gonçalves MA, Meira W Jr (2010) Learning to rank for content-based image retrieval. In: MIR’10, pp 285–294 Fox EA, Shaw JA (1994) Combination of multiple searches. In: The second text retrieval conference (TREC-2), NIST, NIST Special Publication, vol 500–215, pp 243–252 Gopalan R, Turaga P, Chellappa R (2010) Articulation-invariant representation of non-planar shapes. In: Proceedings of the 11th European conference on computer vision: part III, ECCV’10, pp 286–299 Huang J, Kumar SR, Mitra M, Zhu WJ, Zabih R (1997) Image indexing using color correlograms. In: CVPR’97, p 762 Jégou H, Harzallah H, Schmid C (2007) A contextual dissimilarity measure for accurate and efficient image search. In: CVPR, pp 1–8 Ji S, Zhou K, Liao C, Zheng Z, Xue GR, Chapelle O, Sun G, Zha H (2009) Global ranking by exploiting user clicks. In: SIGIR’09, pp 35–42 Kontschieder P, Donoser M, Bischof H (2009) Beyond pairwise shape similarity analysis. ACCV’09, pp 655–666 Kovalev V, Volmer S (1998) Color co-occurence descriptors for querying-by-example. In: MMM’98, p 32 Latecki LJ, Lakmper R, Eckhardt U (2000) Shape descriptors for non-rigid shapes with a single closed contour. In: CVPR, pp 424–429 Ling H, Jacobs DW (2007) Shape classification using the inner-distance. PAMI 29(2):286–299. doi:10.1109/TPAMI.2007.41 Ling H, Yang X, Latecki LJ (2010) Balancing deformability and discriminability for shape matching. ECCV 3:411–424 Liu YT, Liu TY, Qin T, Ma ZM, Li H (2007) Supervised rank aggregation. In: WWW 2007, pp 481–490 Ojala T, Pietikäinen M, Mäenpää T (2002) Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. PAMI 24(7):971–987 Park G, Baek Y, Lee HK (2005) Re-ranking algorithm using post-retrieval clustering for content-based image retrieval. Inf Process Manag 41(2):177–194 Pedronette DCG, da S Torres R (2010) Exploiting contextual information for image re-ranking. CIARP 1:541–548 Pedronette DCG, da S Torres R (2010) Shape retrieval using contour features and distance optmization. In: VISAPP, vol 1, pp 197–202 Pedronette DCG, da S Torres R (2011) Exploiting clustering approaches for image re-ranking. J Vis Languages Comput 22(6):453–466 Pedronette DCG, da S Torres R (2011) Exploiting contextual information for rank aggregation. ICIP, pp 97–100 Perronnin F, Liu Y, Renders JM (2009) A family of contextual measures of similarity between distributions with application to image retrieval. CVPR, pp 2358–2365 Qin T, Liu TY, Zhang XD, Wang DS, Li H (2008) Global ranking using continuous conditional random fields. In: NIPS, pp 1281–1288 Schalekamp F, Zuylen A (1998) Rank aggregation: together were strong. In: Proceeding. of 11th ALENEX, pp 38–51 Schwander O, Nielsen F (2010) Reranking with contextual dissimilarity measures from representational Bregmanl k-means. In: VISAPP, vol 1, pp 118–122 Sculley D (2007) Rank aggregation for similar items. In: SDM Stehling RO, Nascimento MA, Falcão AX (2002) A compact and efficient image retrieval approach based on border/interior pixel classification. In: CIKM’02, pp 102–109 Swain MJ, Ballard DH (1991) Color indexing. IJCV 7(1):11–32 Tao B, Dickinson BW (2000) Texture recognition and image retrieval using gradient indexing. JVCIR 11(3):327–342 Temlyakov A, Munsell BC, Waggoner JW, Wang S (2010) Two perceptually motivated strategies for shape classification. CVPR 1:2289–2296 Tu Z, Yuille AL (2004) Shape matching and recognition-using generative models and informative features. ECCV, pp 195–209 van de Weijer J, Schmid C (2006) Coloring local feature extraction. In: ECCV, vol Part II. Springer, Berlin, pp 334–348 Wang J, Li Y, Bai X, Zhang Y, Wang C, Tang N (2011) Learning context-sensitive similarity by shortest path propagation. Pattern Recogn 44:2367–2374 Yang L, Ji D, Zhou G, Nie Y, Xiao G (2006) Document re-ranking using cluster validation and label propagation. In: CIKM’06, pp 690–697 Yang X, Koknar-Tezel S, Latecki LJ (2009) Locally constrained diffusion process on locally densified distance spaces with applications to shape retrieval. In: CVPR, pp 357–364 Yang X, Latecki LJ (2011) Affinity learning on a tensor product graph with applications to shape and image retrieval. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 2369–2376 Yang X, Bai X, Latecki LJ, Tu Z (2008) Improving shape retrieval by learning graph transduction. ECCV 5305:788–801 Zhao D, Lin Z, Tang X (2007) Contextual distance for data perception. In: ICCV Zhu X (2005) Semi-supervised learning with graphs. PhD thesis, Pittsburgh, PA, USA, chair-Lafferty, John and Chair-Rosenfeld, Ronald