Unsupervised Segmentation Evaluation for Image Annotation

Annette Morales-González, Edel García-Reyes, Luis Enrique Sucar

2015

Abstract

Unsupervised segmentation evaluation measures are usually validated against human-generated ground-truth. Nevertheless, with the recent growth of image classification methods that use hierarchical segmentation-based representations, it would be desirable to assess the performance of unsupervised segmentation evaluation to select the most suitable levels to perform recognition tasks. Another problem is that unsupervised segmentation evaluation measures use only low-level features, which makes difficult to evaluate how well an object is outlined. In this paper we propose to use four semantic measures, that combined with other state-of-the-art measures improve the evaluation results and also, we validate the results of each unsupervised measure against an image annotation algorithm ground truth, showing that using measures that try to emulate human behaviour is not necessarily what an automatic recognition algorithm may need. We employed the Stanford Background Dataset to validate an image annotation algorithm that includes segmentation evaluation as starting point, and the proposed combination of unsupervised measures showed the best annotation accuracy results.

References

  1. Arbelaez, P., Hariharan, B., Gu, C., Gupta, S., Bourdev, L. D., and Malik, J. (2012). Semantic segmentation using regions and parts. In CVPR, pages 3378-3385. IEEE.
  2. Csurka, G., Larlus, D., and Perronnin, F. (2013). What is a good evaluation measure for semantic segmentation? In 24th British Machine Vision Conference (BMVC), University of Bristol, United Kingdom.
  3. Dogra, D. P., Majumdar, A. K., and Sural, S. (2012). Evaluation of segmentation techniques using region area and boundary matching information. J. Vis. Comun. Image Represent., 23(1):150-160.
  4. Gould, S., Fulton, R., and Koller, D. (2009). Decomposing a scene into geometric and semantically consistent regions. In ICCV, pages 1-8. IEEE.
  5. Haxhimusa, Y. and Kropatsch, W. G. (2004). Segmentation graph hierarchies. In Proceedings of Joint International Workshops on Structural, Syntactic, and Statistical Pattern Recognition S+SSPR 2004, volume LNCS 3138, pages 343-351. Springer, Berlin Heidelberg, New York.
  6. Huang, Q., Han, M., Wu, B., and Ioffe, S. (2011). A hierarchical conditional random field model for labeling and segmenting images of street scenes. 2013 IEEE Conference on Computer Vision and Pattern Recognition, 0:1953-1960.
  7. Khan, J. F. and Bhuiyan, S. M. (2014). Weighted entropy for segmentation evaluation. Optics and Laser Technology, 57(0):236 - 242. Optical Image Processing.
  8. Morales-González, A. and García-Reyes, E. B. (2013). Simple object recognition based on spatial relations and visual features represented using irregular pyramids. Multimedia Tools Appl., 63(3):875-897.
  9. Morales-González, A., Reyes, E. B. G., and Sucar, L. E. (2013). Improving image segmentation for boosting image annotation with irregular pyramids. In CIARP (1), volume 8258 of LNCS, pages 399-406. Springer.
  10. Olson, C. R. (2001). Object-based vision and attention in primates. Current Opinion in Neurob., 11:171-179.
  11. Russakovsky, O., Deng, J., Krause, J., Berg, A., and Li, F. (2014). Results of ILSVRC2013. http://www.imagenet.org/challenges/LSVRC/2013/results.php.
  12. Russell, C., Ladicky, L., Kohli, P., and Torr, P. H. S. (2014). Associative hierarchical random fields. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36(6):1-1.
  13. Song, Y.-Z., Arbelaez, P., Hall, P. M., Li, C., and Balikai, A. (2010). Finding semantic structures in image hierarchies using laplacian graph energy. In ECCV (4), volume 6314 of LNCS, pages 694-707. Springer.
  14. van de Sande, K. E. A., Uijlings, J. R. R., Gevers, T., and Smeulders, A. W. M. (2011). Segmentation as selective search for object recognition. In Proceedings of ICCV 7811, pages 1879-1886. IEEE Computer Society.
  15. Yang, P., Liu, W., Zhou, B. B., Chawla, S., and Zomaya, A. Y. (2013). Ensemble-based wrapper methods for feature selection and class imbalance learning. In PAKDD (1), volume 7818 of LNCS, pages 544-555. Springer.
  16. Zankl, G., Haxhimusa, Y., and Ion, A. (2012). Interactive labeling of image segmentation hierarchies. In DAGM/OAGM Symposium, volume 7476 of LNCS, pages 11-20. Springer.
  17. Zhang, H., Fritts, J. E., and Goldman, S. A. (2008). Image segmentation evaluation: A survey of unsupervised methods. Comput. Vis. Image Underst., 110(2):260- 280.
  18. Zhang, S. and Xie, M. (2013). Beyond sliding windows: Object detection based on hierarchical segmentation model. In International Conference on Communications, Circuits and Systems (ICCCAS), pages 263 - 266. IEEE.
Download


Paper Citation


in Harvard Style

Morales-González A., García-Reyes E. and Sucar L. (2015). Unsupervised Segmentation Evaluation for Image Annotation . In Proceedings of the 10th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2015) ISBN 978-989-758-090-1, pages 148-155. DOI: 10.5220/0005314201480155


in Bibtex Style

@conference{visapp15,
author={Annette Morales-González and Edel García-Reyes and Luis Enrique Sucar},
title={Unsupervised Segmentation Evaluation for Image Annotation},
booktitle={Proceedings of the 10th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2015)},
year={2015},
pages={148-155},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005314201480155},
isbn={978-989-758-090-1},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 10th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2015)
TI - Unsupervised Segmentation Evaluation for Image Annotation
SN - 978-989-758-090-1
AU - Morales-González A.
AU - García-Reyes E.
AU - Sucar L.
PY - 2015
SP - 148
EP - 155
DO - 10.5220/0005314201480155