Advertisement

International Journal of Computer Vision

, Volume 87, Issue 3, pp 284–303 | Cite as

From Images to Shape Models for Object Detection

  • Vittorio Ferrari
  • Frederic Jurie
  • Cordelia Schmid
Article

Abstract

We present an object class detection approach which fully integrates the complementary strengths offered by shape matchers. Like an object detector, it can learn class models directly from images, and can localize novel instances in the presence of intra-class variations, clutter, and scale changes. Like a shape matcher, it finds the boundaries of objects, rather than just their bounding-boxes. This is achieved by a novel technique for learning a shape model of an object class given images of example instances. Furthermore, we also integrate Hough-style voting with a non-rigid point matching algorithm to localize the model in cluttered images. As demonstrated by an extensive evaluation, our method can localize object boundaries accurately and does not need segmented examples for training (only bounding-boxes).

Keywords

Object class detection Learning category models Local contour features Shape matching 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Basri, R., Costa, L., Geiger, D., & Jacobs, D. (1998). Determining the similarity of deformable shapes. Vision Research, 38, 2365–2385. CrossRefGoogle Scholar
  2. Belongie, S., & Malik, J. (2002). Shape matching and object recognition using shape contexts. Pattern Analysis and Machine Intelligence, 24(4), 509–522. CrossRefGoogle Scholar
  3. Berg, A., Berg, T., & Malik, J. (2005). Shape matching and object recognition using low distortion correspondence, CVPR. Google Scholar
  4. Borenstein, E., & Ullman, S. (2002). Class-specific, top-down segmentation, ECCV. Google Scholar
  5. Chui, H., & Rangarajan, A. (2003). A new point matching algorithm for non-rigid registration. CVIU, 89(2–3), 114–141. zbMATHGoogle Scholar
  6. Chum, O., & Zisserman, A. (2007). An exemplar model for learning object classes, CVPR. Google Scholar
  7. Cootes, T. (2000). An introduction to active shape models. Google Scholar
  8. Cootes, T., Taylor, C., Cooper, D., & Graham, J. (1995). Active shape models: Their training and application. CVIU, 61(1), 38–59. Google Scholar
  9. Cremers, D., Kohlberger, T., & Schnorr, C. (2002). Nonlinear shape statistics in Mumford-Shah based segmentation, ECCV. Google Scholar
  10. Dalal, N., & Triggs, B. (2005). Histograms of oriented gradients for human detection, CVPR. Google Scholar
  11. Elidan, G., Heitz, G., & Koller, D. (2006). Learning object shape: From drawings to images, CVPR. Google Scholar
  12. Felzenswalb, P. (2005). Representation and detection of deformable shapes. Pattern Analysis and Machine Intelligence, 27(2), 208–220. CrossRefGoogle Scholar
  13. Fergus, R., Perona, P., & Zisserman, A. (2003). Object class recognition by unsupervised scale-invariant learning, CVPR. Google Scholar
  14. Ferrari, V., Tuytelaars, T., & van Gool, L. (2004). Simultaneous object recognition and segmentation by image exploration, ECCV. Google Scholar
  15. Ferrari, V., Tuytelaars, T., & Van Gool, L. (2006). Object detection with contour segment networks, ECCV. Google Scholar
  16. Ferrari, V., Jurie, F., & Schmid, C. (2007). Accurate object detection with deformable shape models learnt from images, CVPR. Google Scholar
  17. Ferrari, V., Fevrier, L., Jurie, F., & Schmid, C. (2008). Groups of adjacent contour segments for object detection. Pattern Analysis and Machine Intelligence, 30(1), 36–51. CrossRefGoogle Scholar
  18. Fritz, M., & Schiele, B. (2006). Towards unsupervised discovery of visual categories, DAGM. Google Scholar
  19. Fritz, M., & Schiele, B. (2008). Decomposition, discovery and detection of visual categories using topic models, CVPR. Google Scholar
  20. Gavrila, D. (1998). Multi-feature hierarchical template matching using distance transforms, ICPR. Google Scholar
  21. Gdalyahu, Y., & Weinshall, D. (1999). Flexible syntactic matching of curves and its application to automatic hierarchical classification of silhouettes. Pattern Analysis and Machine Intelligence, 21(12), 1312–1328. CrossRefGoogle Scholar
  22. Gold, S., & Rangarajan, A. (1996). Graduated assignment algorithm for graph matching. Pattern Analysis and Machine Intelligence, 18(4), 377–388. CrossRefGoogle Scholar
  23. Hill, A., & Taylor, C. (1996). A method of non-rigid correspondence for automatic landmark identification, BMVC. Google Scholar
  24. Jurie, F., & Schmid, C. (2004). Scale-invariant shape features for recognition of object categories, CVPR. Google Scholar
  25. Lamdan, Y., Schwartz, J., & Wolfson, H. (1990). Affine invariant model-based object recognition. IEEE Transactions on Robotics and Automation, 6(5), 578–589. CrossRefGoogle Scholar
  26. Leibe, B., & Schiele, B. (2004). Scale-invariant object categorization using a scale-adaptive mean-shift search, DAGM. Google Scholar
  27. Leordeanu, M., Hebert, M., & Sukthankar, R. (2007). Beyond local appearance: Category recognition from pairwise interactions of simple features, CVPR. Google Scholar
  28. Martin, D., Fowlkes, C., & Malik, J. (2004). Learning to detect natural image boundaries using local brightness, color, and texture cues. Pattern Analysis and Machine Intelligence, 26(5), 530–549. CrossRefGoogle Scholar
  29. Mokhtarian, F., & Mackworth, A. (1986). Scale-based description and recognition of planar curves and two-dimensional shapes. Pattern Analysis and Machine Intelligence, 8(1), 34–43. CrossRefGoogle Scholar
  30. Opelt, A., Pinz, A., & Zisserman, A. (2006). A boundary-fragment-model for object detection, ECCV. Google Scholar
  31. Pentland, A., & Sclaroff, S. (1991). Closed-form solutions for physically based shape modeling and recognition. Pattern Analysis and Machine Intelligence, 13(7), 715–729. CrossRefGoogle Scholar
  32. Quack, T., Ferrari, V., Leibe, B., & Van Gool, L. (2007). Efficient mining of frequent and distinctive feature configurations, ICCV. Google Scholar
  33. Ramanan, D. (2006). Learning to parse images of articulated bodies, NIPS. Google Scholar
  34. Ravishankar, S., Jain, A., & Mittal, A. (2008). Multi-stage contour based detection of deformable objects, ECCV. Google Scholar
  35. Schwartz, J., & Felzenszwalb, P. (2007). Hierarchical matching of deformable shapes, CVPR. Google Scholar
  36. Sharvit, D., Chan, J., Tek, H., & Kimia, B. (1998). Symmetry-based indexing of image databases. IEEE workshop on content-based access of image and video libraries. Google Scholar
  37. Shotton, J., Blake, A., & Cipolla, R. (2005). Contour-based learning for object detection, ICCV. Google Scholar
  38. Sebastian, T., Klein, P., & Kimia, B. (2004). Recognition of shapes by editing their shock graphs. Pattern Analysis and Machine Intelligence, 26(5), 550–571. CrossRefGoogle Scholar
  39. Torralba, A., Murphy, K., & Freeman, W. (2004). Sharing features: Efficient boosting procedures for multiclass object detection, CVPR. Google Scholar
  40. Winn, J., & Jojic, N. (2005). LOCUS: Learning object classes with unsupervised segmentation, ICCV. Google Scholar
  41. Winn, J., & Shotton, J. (2006). The layout consistent random field for recognizing and segmenting partially occluded objects, CVPR. Google Scholar
  42. Zhu, Q., Song, G., & Shi, J. (2007). Untangling cycles for contour grouping, ICCV. Google Scholar
  43. Zhu, Q., Wang, L., Wu, Y., & Shi, J. (2008). Contour context selection for object detection: A set-to-set contour matching approach, ECCV. Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2009

Authors and Affiliations

  • Vittorio Ferrari
    • 1
  • Frederic Jurie
    • 2
  • Cordelia Schmid
    • 3
  1. 1.ETH ZurichZurichSwitzerland
  2. 2.University of CaenCaenFrance
  3. 3.INRIA GrenobleGrenobleFrance

Personalised recommendations