Abstract
While the majority of computer vision systems are based on representing images by local features, the design of the latter has been so far mostly empirical. In this Chapter we propose to tie the design of local features to their systematic evaluation on a realistic ground-truthed dataset. We propose a novel mathematical characterisation of the co-variance properties of the features which accounts for deviation from the usual idealised image affine (de)formation model. We propose novel metrics to evaluate the features and we show how these can be used to automatically design improved features.
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
Preview
Unable to display preview. Download preview PDF.
References
Chum, O., Matas, J.: Matching with PROSAC – progressive sample consensus. In: Proceedings of the International Conference on Computer Vision and Pattern Recognition (2005)
Fraundorfer, F., Bischof, H.: A novel performance evaluation method of local detectors on non-planar scenes. In: Proceedings of the International Conference on Computer Vision and Pattern Recognition (2005)
Hertz, T., Bar-Hillel, A., Weinshall, D.: Learning distance functions for image retrieval. In: Proceedings of the International Conference on Computer Vision and Pattern Recognition (2004)
Lepetit, V., Lagger, P., Fua, P.: Randomized trees for real-time keypoint recognition. In: Proceedings of the International Conference on Computer Vision and Pattern Recognition (2005)
Ling, H., Jacobs, D.W.: Deformation invariant image matching. In: Proceedings of the International Conference on Computer Vision (2005)
Lowe, D.G.: Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision 2(60), 91–110 (2004)
Ma, Y., Soatto, S., Kosecka, J., Sastry, S.S.: An Invitation to 3-D Vision. Springer, Heidelberg (2003b)
Matas, J., Chum, O., Urban, M., Pajdla, T.: Robust wide baseline stereo from maximally stable extremal regions. In: Proceedings of the Brithish Machine Vision Conference (2002)
Mikolajczyk, K., Tuytelaars, T., Schmid, C., Zisserman, A., Matas, J., Schaffalitzky, F., Kadir, T., Van Gool, L.: A comparison of affine region detectors. International Journal of Computer Vision 1(60), 63–86 (2004)
Moreels, P., Perona, P.: Evaluation of features detectors and descriptors based on 3D objects. In: Proceedings of the International Conference on Computer Vision (2005)
Piqueres, J.V.: The persistence of ignorance (2006), http://www.ignorancia.org/
Rockett, P.I.: Performance assesment of feature detection algorithms: A methodology and case study on corner detectors. Transaction on Image Processing 12(12) (2003)
Roth, S., Black, M.J.: On the spatial statistics of optical flow. In: Proceedings of the International Conference on Computer Vision (2005)
Vedaldi, A.: A ground-truthed dataset for evaluation and learning of viewpoint invariant features (2008), http://vision.ucla.edu/gtvpi
Vedaldi, A., Fulkerson, B.: VLFeat: An open and portable library of computer vision algorithms (2008), http://www.vlfeat.org/
Vedaldi, A., Soatto, S.: Features for recognition: Viewpoint invariance for non-planar scenes. In: Proceedings of the International Conference on Computer Vision (2005)
Winder, S.A.J., Brown, M.: Learning local image descriptors. In: Proceedings of the International Conference on Computer Vision and Pattern Recognition (2007)
Zhou, S.K., Shao, J., Georgescu, B., Comaniciu, D.: BoostMotion: Boosting a discriminative similarity function for motion estimation. In: Proceedings of the International Conference on Computer Vision and Pattern Recognition (2006)
Zhuang, H., Sudhakar, R.: Simultaneous rotation and translation fitting of two 3-D point sets. Transaction on Systems, Man, and Cybernetics 27(1) (1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Vedaldi, A., Ling, H., Soatto, S. (2010). Knowing a Good Feature When You See It: Ground Truth and Methodology to Evaluate Local Features for Recognition. In: Cipolla, R., Battiato, S., Farinella, G.M. (eds) Computer Vision. Studies in Computational Intelligence, vol 285. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-12848-6_2
Download citation
DOI: https://doi.org/10.1007/978-3-642-12848-6_2
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-12847-9
Online ISBN: 978-3-642-12848-6
eBook Packages: EngineeringEngineering (R0)