Are Edges Incomplete?

International Journal of Computer Vision - Tập 34 Số 2 - Trang 97-122 - 1999
Elder, James H.

Tóm tắt

We address the problem of computing a general-purpose early visual representation that satisfies two criteria. 1) Explicitness: To be more useful than the original pixel array, the representation must take a significant step toward making important image structure explicit. 2) Completeness: To support a diverse set of high-level tasks, the representation must not discard information of potential perceptual relevance. The most prevalent representation in image processing and computer vision that satisfies the completeness criterion is the wavelet code. In this paper, we propose a very different code which represents the location of each edge and the magnitude and blur scale of the underlying intensity change. By making edge structure explicit, we argue that this representation better satisfies the first criterion than do wavelet codes. To address the second criterion, we study the question of how much visual information is lost in the representation. We report a novel method for inverting the edge code to reconstruct a perceptually accurate estimate of the original image, and thus demonstrate that the proposed representation embodies virtually all of the perceptually relevant information contained in a natural image. This result bears on recent claims that edge representations do not contain all of the information needed for higher level tasks.

Từ khóa


Tài liệu tham khảo

citation_title=The plenoptic function and the elements of early vision; citation_inbook_title=Computational Models of Visual Processing; citation_publication_date=1991; citation_id=CR1; citation_author=E. Adelson; citation_publisher=MIT Press

citation_journal_title=Proc. of SPIE; citation_title=Orthogonal pyramid transforms for image coding; citation_author=E. Adelson, E. Simoncelli, R. Hingorani; citation_volume=845; citation_publication_date=1987; citation_pages=50-58; citation_id=CR2

citation_journal_title=IEEE Transactions on Pattern Analysis and Machine Intelligence; citation_title=Face recognition: the problem of compensating for illumination changes; citation_author=Y. Adini, Y. Moses, S. Ullman; citation_volume=19; citation_issue=7; citation_publication_date=1997; citation_pages=721-732; citation_id=CR3

Barlow, H. 1961. The coding of sensory messages. In Current Problems in Animal Behavior, W. Thorpe and O. Zangwill (Eds.), Cambridge U. Press, pp. 331–360.

citation_journal_title=Proc. IEEE; citation_title=Computational vision; citation_author=H. Barrow, J. Tenenbaum; citation_volume=69; citation_publication_date=1981; citation_pages=572-595; citation_id=CR5

citation_journal_title=CVGIP: Graphical Models and Image Processing; citation_title=Image encoding, labeling and reconstruction from differential geometry; citation_author=E. Barth, T. Caelli, C. Zetzsche; citation_volume=55; citation_issue=6; citation_publication_date=1993; citation_pages=428-446; citation_id=CR6

citation_title=What is the set of images of an object under all possible lighting conditions?; citation_inbook_title=Proc. IEEE Conf. Computer Vision Pattern Recognition; citation_publication_date=1996; citation_pages=270-277; citation_id=CR7; citation_author=P. Belhumeur; citation_author=D. Kriegman; citation_publisher=IEEE Computer Society, IEEE Computer Society Press

Canny, J. 1983. Finding edges and lines in images. Master's thesis, MIT Artificial Intelligence Laboratory.

Carlsson, S. 1984. Sketch based image coding. In Proc. of Premier Colloque Image, Biarritz, France, pp. 71–77.

citation_journal_title=Signal Processing; citation_title=Sketch based coding of grey level images; citation_author=S. Carlsson; citation_volume=15; citation_publication_date=1988; citation_pages=57-83; citation_id=CR10

Cox, I., Boie, R., and Wallach, D. 1990. Line recognition. In Proc. Int. Conf. on Pattern Recognition, Atlantic City, NJ, pp. 639–645.

citation_journal_title=Int. J. Comp. Vision; citation_title=A Bayesian multiple-hypothesis approach to edge grouping and contour segmentation; citation_author=I. Cox, J. Rehg, S. Hingorani; citation_volume=11; citation_issue=1; citation_publication_date=1993; citation_pages=5-24; citation_id=CR12

citation_journal_title=CVGIP: Graphical Models and Image Processing; citation_title=An Edge-Based Description of Color Images; citation_author=A. Cumani, P. Grattoni, A. Guiducci; citation_volume=53; citation_issue=4; citation_publication_date=1991; citation_pages=313-323; citation_id=CR13

citation_journal_title=IEEE Trans. Acoust., Speech, Signal Processing; citation_title=Reconstruction of nonperiodic two-dimensional signals from zero crossings; citation_author=S. Curtis, S. Shitz, A. Oppenheim; citation_volume=35; citation_publication_date=1987; citation_pages=890-893; citation_id=CR14

Daubechies, I. 1991. Ten lectures on wavelets. In CBMS-NSF Series Appl. Math., SIAM.

citation_journal_title=Int. J. Computer Vision; citation_title=Potentials, valleys and dynamic global coverings; citation_author=C. David, S. Zucker; citation_volume=5; citation_publication_date=1990; citation_pages=219-238; citation_id=CR16

citation_journal_title=Int. J. Computer Vision; citation_title=Using Canny's criteria to derive a recursively implemented optimal edge detector; citation_author=R. Deriche; citation_volume=1; citation_issue=2; citation_publication_date=1987; citation_pages=167-187; citation_id=CR17

citation_journal_title=Int. J. Comp. Vision; citation_title=The multiscale veto model: a two-stage analog network for edge detection and image reconstruction; citation_author=L. Dron; citation_volume=56; citation_publication_date=1977; citation_pages=487-510; citation_id=CR18

citation_journal_title=European Conf. on Visual Perception; citation_title=Brightness filling-in of natural images; citation_author=J. Elder; citation_volume=26; citation_issue=Suppl.; citation_publication_date=1997; citation_pages=57; citation_id=CR19

citation_journal_title=J. Invest. Opthalm. Visual Sci.; citation_title=Edge classification in natural images; citation_author=J. Elder, D. Beniaminov, G. Pintilie; citation_volume=40; citation_issue=4; citation_publication_date=1999; citation_pages=1897; citation_id=CR20

citation_title=Interactive contour editing; citation_inbook_title=Proc. IEEE Conf. Computer Vision Pattern Recognition; citation_publication_date=1998; citation_pages=374-381; citation_id=CR21; citation_author=J. Elder; citation_author=R. Goldberg; citation_publisher=IEEE Computer Society, IEEE Computer Society Press

citation_journal_title=J. Invest. Opthalm. Visual Sci.; citation_title=Rapid processing of cast and attached shadows; citation_author=J. Elder, S. Trithart, G. Pintilie, D. MacLean; citation_volume=39; citation_issue=4; citation_publication_date=1998; citation_pages=S853; citation_id=CR22

citation_journal_title=J. Invest. Opthalm. Visual Sci.; citation_title=The local character of generalized luminance transitions; citation_author=J. Elder, S. Zucker; citation_volume=36; citation_issue=4; citation_publication_date=1995; citation_pages=S836; citation_id=CR23

citation_journal_title=IEEE Pattern Anal. Machine Intell.; citation_title=Local scale control for edge detection and blur estimation; citation_author=J. Elder, S. Zucker; citation_volume=20; citation_issue=7; citation_publication_date=1998; citation_pages=699-716; citation_id=CR24

citation_title=Computing contour closure; citation_inbook_title=Proc 4th European Conf. on Computer Vision; citation_publication_date=1996; citation_pages=399-412; citation_id=CR25; citation_author=J. Elder; citation_author=S. Zucker; citation_publisher=Springer Verlag

citation_title=Local scale control for edge detection and blur estimation; citation_inbook_title=Proc. 4th European Conf. on Computer Vision; citation_publication_date=1996; citation_pages=57-69; citation_id=CR26; citation_author=J. Elder; citation_author=S. Zucker; citation_publisher=Springer Verlag

citation_title=Scale space localization, blur and contour-based image coding; citation_inbook_title=Proc. IEEE Conf. Computer Vision Pattern Recognition; citation_publication_date=1996; citation_pages=27-34; citation_id=CR27; citation_author=J. Elder; citation_author=S. Zucker; citation_publisher=IEEE Computer Society, IEEE Computer Society Press

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=Investigation of methods for determining depth from focus; citation_author=J. Ens, P. Lawrence; citation_volume=15; citation_issue=2; citation_publication_date=1993; citation_pages=97-108; citation_id=CR28

citation_journal_title=Computer Graphics and Image Processing; citation_title=Detection of roads and linear structures in low-resolution aerial imagery using a multi-source knowledge integration technique; citation_author=M. Fischler, J. Tenenbaum, H. Wolf; citation_volume=18; citation_issue=4; citation_publication_date=1981; citation_pages=201-223; citation_id=CR29

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=The design and use of steerable filters; citation_author=W. Freeman, E. Adelson; citation_volume=13; citation_issue=9; citation_publication_date=1991; citation_pages=891-906; citation_id=CR30

citation_journal_title=Pattern Recognition Letters; citation_title=Contour Coding for Image Description; citation_author=P. Grattoni, A. Guiducci; citation_volume=11; citation_publication_date=1990; citation_pages=95-105; citation_id=CR31

citation_journal_title=IEEE Trans. on Acoustics, Speech, and Signal Processing; citation_title=Reconstructions from zero crossings in scale space; citation_author=R. Hummel, R. Moniot; citation_volume=37; citation_issue=12; citation_publication_date=1989; citation_pages=2111-2130; citation_id=CR32

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=Logical/linear operators for image curves; citation_author=L. Iverson, S. Zucker; citation_volume=17; citation_issue=10; citation_publication_date=1995; citation_pages=982-996; citation_id=CR33

citation_journal_title=Perception; citation_title=Moving cast shadows induce apparent motion in depth; citation_author=D. Kersten, P. Mamassian, D. Knill; citation_volume=26; citation_issue=2; citation_publication_date=1997; citation_pages=171-192; citation_id=CR34

citation_journal_title=Biol. Cybern.; citation_title=The structure of images; citation_author=J. Koenderink; citation_volume=50; citation_publication_date=1984; citation_pages=363-370; citation_id=CR35

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=The local structure of image discontinuities in one dimension; citation_author=Y. Leclerc, S. Zucker; citation_volume=9; citation_issue=3; citation_publication_date=1987; citation_pages=341-355; citation_id=CR36

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=Scale-space for discrete signals; citation_author=T. Lindeberg; citation_volume=12; citation_issue=3; citation_publication_date=1990; citation_pages=234-254; citation_id=CR37

citation_title=Edge detection and ridge detection with automatic scale selection; citation_inbook_title=IEEE Conf. Computer Vision Pattern Recognition; citation_publication_date=1996; citation_pages=465-470; citation_id=CR38; citation_author=T. Lindeberg; citation_publisher=IEEE Computer Society, IEEE Computer Society Press

citation_journal_title=Bell Syst. Tech. J.; citation_title=Information in the zero-crossings of bandpass signals; citation_author=B.F. Logan; citation_volume=56; citation_publication_date=1977; citation_pages=487-510; citation_id=CR39

citation_journal_title=IEEE Trans. Inform. Theory; citation_title=Multifrequency channel decompositions of images and wavelet models; citation_author=S. Mallat; citation_volume=37; citation_issue=12; citation_publication_date=1989; citation_pages=2091-2110; citation_id=CR40

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=Characterization of signals from multiscale edges; citation_author=S. Mallat, S. Zhong; citation_volume=14; citation_publication_date=1992; citation_pages=710-732; citation_id=CR41

citation_title=Vision; citation_publication_date=1982; citation_id=CR42; citation_author=D. Marr; citation_publisher=W.H. Freeman

citation_journal_title=Proc. R. Soc. Lond. B; citation_title=Theory of edge detection; citation_author=D. Marr, E. Hildreth; citation_volume=207; citation_publication_date=1980; citation_pages=187-217; citation_id=CR43

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=Shape from focus; citation_author=S. Nayar, N. Yasuo; citation_volume=16; citation_issue=8; citation_publication_date=1994; citation_pages=824-831; citation_id=CR44

citation_journal_title=Vision Res.; citation_title=Brightness perception and filling-in; citation_author=M. Paradiso, K. Nakayama; citation_volume=7/8; citation_publication_date=1991; citation_pages=1221-1236; citation_id=CR45

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=A new sense for depth of field; citation_author=A. Pentland; citation_volume=9; citation_issue=4; citation_publication_date=1987; citation_pages=523-531; citation_id=CR46

citation_journal_title=IEEE Trans. Pattern Anal. Machine Intell.; citation_title=Deformable kernels for early vision; citation_author=P. Perona; citation_volume=17; citation_issue=5; citation_publication_date=1995; citation_pages=488-499; citation_id=CR47

Press, W., Teukolsky, S., Vetterling, W., and Flannery, B. 1992. Numerical Recipes in C (2 edition)., Cambridge University Press, chap 19, pp. 871–882.

citation_title=Machine perception of 3-dimensional solids; citation_inbook_title=Optical and Electro-Optical Information Processing; citation_publication_date=1965; citation_id=CR49; citation_author=L. Roberts; citation_publisher=MIT Press

citation_title=Structural saliency: the detection of globally salient structures using a locally connected network; citation_inbook_title=Proc. 2nd Int. Conf. on Computer Vision; citation_publication_date=1988; citation_pages=321-327; citation_id=CR50; citation_author=A. Sha'ashua; citation_author=S. Ullman; citation_publisher=IEEE Computer Soc. Press

citation_journal_title=IEEE Trans. on Inf. Theory; citation_title=Shiftable multiscale transforms; citation_author=E. Simoncelli, W. Freeman, E. Adelson, D. Heeger; citation_volume=38; citation_issue=2; citation_publication_date=1992; citation_pages=587-607; citation_id=CR51

citation_journal_title=Signal Processing; citation_title=Multidimensional subband coding: some theory and algorithms; citation_author=M. Vetterli; citation_volume=6; citation_issue=2; citation_publication_date=1984; citation_pages=97-112; citation_id=CR52

citation_journal_title=Amer. J. Psychol.; citation_title=Studies on contour: I. Qualitative analyses; citation_author=H. Werner; citation_volume=47; citation_publication_date=1935; citation_pages=40-64; citation_id=CR53

Witkin, A. 1983. Scale space filtering. In Proc. Int. Joint Conf. on Artif. Intell., Karlsruhe, pp. 1019–1021.

citation_journal_title=J. Opt. Soc. Am. A; citation_title=Fingerprints theorems for zero crossings; citation_author=A. Yuille, T. Poggio; citation_volume=2; citation_issue=5; citation_publication_date=1985; citation_pages=683-692; citation_id=CR55

citation_journal_title=IEEE Trans. Acoust., Speech, Signal Processing; citation_title=Image reconstruction from zero crossings; citation_author=Y. Zeevi, D. Rotem; citation_volume=34; citation_publication_date=1986; citation_pages=1269-1277; citation_id=CR56

Zucker, S. 1986. Early vision. In The Encyclopedia of Artificial Intelligence, S. Shapiro (Ed.), John Wiley.

citation_journal_title=IEEE Trans. Comput.; citation_title=An application of relaxation labeling to line and curve enhancement; citation_author=S. Zucker, R. Hummel, A. Rosenfeld; citation_volume=26; citation_publication_date=1977; citation_pages=394-403; citation_id=CR58