Abstract
Cosegmentation methods segment multiple related images jointly, exploiting their shared appearance to generate more robust foreground models. While existing approaches assume that an oracle will specify which pairs of images are amenable to cosegmentation, in many scenarios such external information may be difficult to obtain. This is problematic, since coupling the “wrong” images for segmentation—even images of the same object class—can actually deteriorate performance relative to single-image segmentation. Rather than manually specify partner images for cosegmentation, we propose to automatically predict which images will cosegment well together. We develop a learning-to-rank approach that identifies good partners, based on paired descriptors capturing the images’ amenability to joint segmentation. We compare our approach to alternative methods for partnering images, including basic image similarity, and show the advantages on two challenging datasets.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We use the terms cosegmentation, joint segmentation, and weakly supervised segmentation interchangeably.
- 2.
Alternatively, one could use regression. However, ranking has the advantage of giving us more control over which training tuples are enforced, and it places emphasis only on the relative scores (not absolute values), which is what we care about for deciding which partner is best.
- 3.
- 4.
- 5.
References
Rother, C., Minka, T., Blake, A., Kolmogorov, V.: Cosegmentation of image pairs by histogram matching - incorporating a global constraint into MRFs. In: CVPR (2006)
Winn, J., Jojic, N.: LOCUS: learning object classes with unsupervised segmentation. In: ICCV (2005)
Alexe, B., Deselaers, T., Ferrari, V.: ClassCut for unsupervised class segmentation. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part V. LNCS, vol. 6315, pp. 380–393. Springer, Heidelberg (2010)
Vicente, S., Rother, C., Kolmogorov, V.: Object cosegmentation. In: CVPR (2011)
Joulin, A., Bach, F., Ponce, J.: Discriminative clustering for image co-segmentation. In: CVPR (2010)
Joulin, A., Bach, F., Ponce, J.: Multi-class cosegmentation. In: CVPR (2012)
Kim, G., Xing, E., Fei Fei, L., Kanade, T.: Distributed cosegmentation via submodular optimization on anisotropic diffusion. In: ICCV (2011)
Rubinstein, M., Joulin, A., Kopf, J., Liu, C.: Unsupervised joint object discovery and segmentation in internet images. In: CVPR (2013)
Cao, L., Fei-Fei, L.: Spatially coherent latent topic model for concurrent segmentation and Classification of objects and scenes. In: ICCV (2007)
Todorovic, S., Ahuja, N.: Unsupervised category modeling, recognition, and segmentation in images. PAMI 30, 2158–2174 (2008)
Lee, Y.J., Grauman, K.: Collect-cut: segmentation with top-down cues discovered in multi-object images. In: CVPR (2010)
Hochbaum, D., Singh, V.: An efficient algorithm for co-segmentation. In: ICCV (2009)
Batra, D., Kowdle, A., Parikh, D., Luo, J., Chen, T.: iCoseg: interactive co-segmentation with intelligent scribble guidance. In: CVPR (2010)
Russell, B., Efros, A., Sivic, J., Freeman, W., Zisserman, A.: Using multiple segmentations to discover objects and their extent in image collections. In: CVPR (2006)
Faktor, A., Irani, M.: “Clustering by composition” – unsupervised discovery of image categories. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part VII. LNCS, vol. 7578, pp. 474–487. Springer, Heidelberg (2012)
Liu, D., Xiong, Y., Pulli, K., Shapiro, L.: Estimating image segmentation difficulty. In: Machine Learning and Data Mining in Pattern Recognition (2011)
Carreira, J., Sminchisescu, C.: CPMC: automatic object segmentation using constrained parametric min-cuts. PAMI 34, 1312–1328 (2012)
Endres, I., Hoiem, D.: Category independent object proposals. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part V. LNCS, vol. 6315, pp. 575–588. Springer, Heidelberg (2010)
Ren, X., Malik, J.: Learning a classication model for segmentation. In: ICCV (2003)
Jain, S., Grauman, K.: Predicting sufficient annotation strength for interactive foreground segmentation. In: ICCV (2013)
Jiang, B., Zhang, L., Lu, H., Yang, C., Yang, M.H.: Saliency detection via absorbing markov chain. In: ICCV (2013)
Rother, C., Kolmogorov, V., Blake, A.: Grabcut -interactive foreground extraction using iterated graph cuts. In: SIGGRAPH (2004)
Kim, J., Liu, C., Sha, F., Grauman, K.: Deformable spatial pyramid matching for fast dense correspondences. In: CVPR (2013)
Joachims, T.: Optimizing search engines with clickthrough data. In: KDD (2002)
Kohli, P., Torr, P.H.S.: Measuring uncertainty in graph cut solutions. CVIU 112, 30–38 (2008)
Torralba, A.: Contextual priming for object detection. Int. J. Comput. Vis. 53, 169–191 (2003)
Kim, J., Grauman, K.: Boundary preserving dense local regions. In: CVPR (2011)
Acknowledgements
This research is supported in part by ONR award N00014-12-1-0068.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Jain, S.D., Grauman, K. (2015). Which Image Pairs Will Cosegment Well? Predicting Partners for Cosegmentation. In: Cremers, D., Reid, I., Saito, H., Yang, MH. (eds) Computer Vision -- ACCV 2014. ACCV 2014. Lecture Notes in Computer Science(), vol 9005. Springer, Cham. https://doi.org/10.1007/978-3-319-16811-1_12
Download citation
DOI: https://doi.org/10.1007/978-3-319-16811-1_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-16810-4
Online ISBN: 978-3-319-16811-1
eBook Packages: Computer ScienceComputer Science (R0)