Visual recognition using local appearance
This paper presents a technique for visual recognition in which physical objects are represented by families of surfaces in a local appearance space. An orthogonal family of local appearance descriptors is obtained by applying principal components analysis to image neighborhoods. The principal components with the largest variance are used to define a space for describing local appearance. The projection of the set of all neighborhoods from an image gives a discrete sampling of a surface in this space. Projecting neighborhoods from images taken at different viewing positions gives a family of surfaces which represent the possible local appearances from those viewing directions.
Recognition is achieved by projecting windows from newly acquired images into the local appearance space and associating them to nearby surfaces. An efficient search tree data structure is used to associate projected points to surfaces. Our results show that in many common situations, a single window is sufficient to obtain the correct recognition. Robust recognition is easily obtained by reinforcing matching using multiple windows and their mutual spatial coherence.
KeywordsRecognition Rate Local Descriptor Visual Recognition Correct Recognition Descriptor Space
Unable to display preview. Download preview PDF.
- 1.H. Bischof A. Leonardis and R. Ebensberger. Robust recognition using eigimages. Technical Report PRIP-TR-47, Pattern Recogntion and Image Processing group — Vienna University of Technology, June 1997.Google Scholar
- 2.M.J. Black and A.D. Jepson. Eigentracking: Robust matching and tracking of articulated objects using a view-based representation”. In Springer Verlag, editor, ECCV'96, Fourth European Conference on Computer Vision, pages 329–342, 1996.Google Scholar
- 3.P. Bobet. TÊte stéréoscopique, Réflexes oculaires et Vision. PhD thesis, INPG France, 1995.Google Scholar
- 5.R.O. Duda and P.E. Hart. Pattern Classification and Scene Analysis, chapter 4. J. Wiley and sons, 1973.Google Scholar
- 6.O. Faugeras. Three-Dimensional Computer Vision: A Geometric Viewpoint. The MIT Press, 1993.Google Scholar
- 8.K. Fukunaga. Statistical Pattern Recognition, chapter Feature Extraction and Linear Mapping for Signal Representation. Academic Press, School of Electrical Engineering, West Lafayette, Indiana, 1990.Google Scholar
- 10.C. Harris and M. Stephens. A combined corner and edge detector. In Proc. 4th Alvey Vision Conference, 1988.Google Scholar
- 11.A. Lux and B. Zoppis. An Experimental Multi-language Environment for the Development of Intelligent Robot Systems. In 5th International Symposium on Intelligent Robotic Systems, SIRS'97, pages 169–174, 1997. more informations at http://pandora.imag.fr/Prima/Ravi/.Google Scholar
- 12.J. Martin and J.L. Crowley. Comparison of correlation techniques. In U. Rembold et al., editor, Intelligent Autonomous Systems — IAS-4, pages 86–93, Karlsruhe, Germany, March 27–30 1995.Google Scholar
- 13.B.W. Mel. Seemore: A view-based approach to 3-d object recognition using multiple visual cues. Advances in neural information processing systems, 8:865–871, 1996.Google Scholar
- 15.S. A. Nene, S. K. Nayar, and H. Murase. Columbia object image library (coil-100). Technical report, Columbia University, New York, february 1996.Google Scholar
- 16.K. Ohba and K. Ikeuchi. Recognition of the multi specularity objects for binpicking task. IROS 96, 3:1440–1448, 1996.Google Scholar
- 17.R. P. N. Rao and D. H. Ballard. Object indexing using an iconic sparse distributed memory. In ICCV'95 Fifth International Conference on Computer Vision, pages 24–31, 1995.Google Scholar
- 18.B. Schiele. Object Recognition using Multidimensional Receptive Field Histograms. PhD thesis, I.N.P. Grenoble, France, 1997. English translation.Google Scholar
- 19.B. Schiele and J. L. Crowley. Object recognition using multidimensional receptive field histograms. In ECCV'96, Fourth European Conference on Computer Vision, Volume I, pages 610–619, 14–16 April 1996.Google Scholar
- 20.C. Schmid and R. Mohr. Combining grayvalue invariants with local constraints for object recognition. In International Conference on Computer Vision and Pattern Recognition, 1996.Google Scholar