More Web Proxy on the site http://driver.im/

Article

Connecting missing links: object discovery from sparse observations using 5 million product images

Authors:

Martial Hebert,

Alexei A. Efros,

Takeo KanadeAuthors Info & Claims

ECCV'12: Proceedings of the 12th European conference on Computer Vision - Volume Part VI

Pages 794 - 807

https://doi.org/10.1007/978-3-642-33783-3_57

Published: 07 October 2012 Publication History

Abstract

Object discovery algorithms group together image regions that originate from the same object. This process is effective when the input collection of images contains a large number of densely sampled views of each object, thereby creating strong connections between nearby views. However, existing approaches are less effective when the input data only provide sparse coverage of object views.

We propose an approach for object discovery that addresses this problem. We collect a database of about 5 million product images that capture 1.2 million objects from multiple views. We represent each region in the input image by a "bag" of database object regions. We group input regions together if they share similar "bags of regions." Our approach can correctly discover links between regions of the same object even if they are captured from dramatically different viewpoints. With the help from these added links, our proposed approach can robustly discover object instances even with sparse coverage of the viewpoints.

References

[1]

Tuytelaars, T., Lampert, C.H., Blaschko, M.B., Buntine, W.: Unsupervised object discovery: A comparison. IJCV (2009).

Digital Library

[2]

Sivic, J., Russell, B.C., Efros, A.A., Zisserman, A., Freeman, W.T.: Discovering object categories in image collections. In: ICCV (2005).

[3]

Russell, B.C., Efros, A.A., Sivic, J., Freeman, W.T., Zisserman, A.: Using multiple segmentations to discover objects and their extent in image collections. In: CVPR (2006).

Digital Library

[4]

Quack, T., Ferrari, V., Leibe, B., Van Gool, L.: Efficient mining of frequent and distinctive feature configurations. In: ICCV 2007 (2007).

[5]

Kim, G., Faloutsos, C., Hebert, M.: Unsupervised modeling of object categories using link analysis techniques. In: CVPR (2008).

[6]

Cho, M., Shin, Y.M., Lee, K.M.: Unsupervised detection and segmentation of identical objects. In: CVPR (2010).

[7]

Lee, Y.J., Grauman, K.: Object-graphs for context-aware category discovery. In: CVPR (2010).

[8]

Payet, N., Todorovic, S.: From a Set of Shapes to Object Discovery. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part V. LNCS, vol. 6315, pp. 57-70. Springer, Heidelberg (2010).

Digital Library

[9]

Kang, H., Hebert, M., Kanade, T.: Discovering object instances from scenes of daily living. In: ICCV (2011).

[10]

Huffman, D.A.: Impossible Objects as Nonsense Sentences. Machine Intelligence 6, 295-323 (1971).

[11]

Li, L.J., Su, H., Xing, E.P., Fei-Fei, L.: Object bank: A high-level image representation for scene classification & semantic feature sparsification. In: NIPS (2010).

[12]

Lim, J., Salakhutdinov, R., Torralba, A.: Transfer learning by borrowing examples. In: NIPS (2011).

[13]

Schroff, F., Treibitz, T., Kriegman, D., Belongie, S.: Pose, illumination and expression invariant pairwise face-similarity measure via doppelgänger list comparison. In: ICCV (2011).

Digital Library

[14]

Torralba, A., Efros, A.A.: Unbiased look at dataset bias. In: CVPR 2011 (2011).

Digital Library

[15]

Ke, Y., Tang, X., Jing, F.: The design of high-level features for photo quality assessment. In: CVPR (2006).

Digital Library

[16]

Felzenszwalb, P.F., Huttenlocher, D.P.: Efficient graph-based image segmentation. IJCV (2004).

Digital Library

[17]

Mishra, A., Aloimonos, Y.: Active segmentation with fixation. In: ICCV (2009).

[18]

Arkin, E.M., Chew, L.P., Huttenlocher, D.P., Kedem, K., Mitchell, J.S.B.: An efficiently computable metric for comparing polygonal shapes. PAMI (1991).

Digital Library

[19]

Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: CVPR (2005).

Digital Library

[20]

Hays, J., Efros, A.A.: Scene completion using millions of photographs. In: SIGGRAPH (2007).

Digital Library

[21]

Torralba, A., Fergus, R., Freeman, W.T.: 80 million tiny images: A large data set for nonparametric object and scene recognition. PAMI (2008).

Digital Library

[22]

Lai, K., Bo, L., Ren, X., Fox, D.: A large-scale hierarchical multi-view rgb-d object dataset. In: ICRA (2011).

[23]

Brown, M., Lowe, D.: Recognising panoramas. In: Proceedings of the 9th International Conference on Computer Vision, Nice, vol. 2, pp. 1218-1225 (2003).

Digital Library

Cited By

Huang JLiu SXing JMei TYan S(2014)Circle & SearchACM Transactions on Multimedia Computing, Communications, and Applications10.1145/263216511:1(1-21)Online publication date: 4-Sep-2014
https://dl.acm.org/doi/10.1145/2632165
Kim GXing ECarterette BDiaz FCastillo CMetzler D(2014)Visualizing brand associations from web community photosProceedings of the 7th ACM international conference on Web search and data mining10.1145/2556195.2556212(623-632)Online publication date: 24-Feb-2014
https://dl.acm.org/doi/10.1145/2556195.2556212

Recommendations

Multiple instance learning with missing object tags
ICIMCS '11: Proceedings of the Third International Conference on Internet Multimedia Computing and Service

In this paper, we have addressed two key issues for leveraging large-scale loosely-tagged images for object classifier training: (a) loose object tags, e.g., multiple object tags are loosely given at the image level without identifying object locations ...
Multi-label multi-instance learning with missing object tags

In this paper, a novel framework is developed for leveraging large-scale loosely tagged images for object classifier training by addressing three key issues jointly: (a) spam tags e.g., some tags are more related to popular query terms rather than the ...
Object recognition and segmentation in videos by connecting heterogeneous visual features

We present an approach for model-free and instance-level object recognition and segmentation in cluttered scenes, based on heterogeneous visual features. The first contribution of this work addresses the description of the visual appearance of objects, ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image Guide Proceedings

ECCV'12: Proceedings of the 12th European conference on Computer Vision - Volume Part VI

October 2012

892 pages

ISBN:9783642337826

Editors:
Andrew Fitzgibbon
Microsoft Research Ltd., Cambridge, UK
,
Svetlana Lazebnik
Dept. of Computer Science, University of North Carolina, Chapel Hill, NC, UK
,
Pietro Perona
Dept. of Computer Science, California Institute of Technology, Pasadena, CA, UK
,
Yoichi Sato
Institute of Industrial Science, The University of Tokyo, Tokyo, CA, Japan
,
Cordelia Schmid
Institute of Industrial Science, INRIA, Montbonnot, CA, France

Sponsors

Adobe
Google Inc.
IBMR: IBM Research
NVIDIA
Microsoft Reasearch: Microsoft Reasearch

Publisher

Springer-Verlag

Berlin, Heidelberg

Publication History

Published: 07 October 2012

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

2
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 01 Mar 2025

Other Metrics

View Author Metrics

Citations

Cited By

Huang JLiu SXing JMei TYan S(2014)Circle & SearchACM Transactions on Multimedia Computing, Communications, and Applications10.1145/263216511:1(1-21)Online publication date: 4-Sep-2014
https://dl.acm.org/doi/10.1145/2632165
Kim GXing ECarterette BDiaz FCastillo CMetzler D(2014)Visualizing brand associations from web community photosProceedings of the 7th ACM international conference on Web search and data mining10.1145/2556195.2556212(623-632)Online publication date: 24-Feb-2014
https://dl.acm.org/doi/10.1145/2556195.2556212

View Options

View options

Figures

Tables

Media

View Table of Conten