18.8 Surface or Contour Motion Using Global Surfaces

Chapter Contents (Back)
Motion, Surface Reconstruction. Surface Reconstruction. See also Structure, Depth, and Shape from Motion.

Murray, D.W., Castelow, D.A., and Buxton, B.F.,
From Image Sequences to Recognized Moving Polyhedral Objects,
IJCV(3), No. 3, September 1989, pp. 181-208.
Springer DOI BibRef 8909
And:
From an Image Sequence to a Recognized Polyhedral Object,
IVC(6), No. 2, May 1988, pp. 107-120.
WWW Link. Use Canny edges and compute motion using 3 frames. Recover the structure and match to CAD-based descriptions. BibRef

Lee, C.H.[Chia-Hoang],
Interpreting Image Curve from Multiframes,
AI(35), No. 2, June 1988, pp. 145-163.
WWW Link. 3-D interpretation of a series of 2-D views of a curve. BibRef 8806

Liu, W., and Kanatani, K.I.,
Interpretation of Conic Motion and Its Applications,
IJCV(10), No. 1, February 1993, pp. 67-84.
Springer DOI Analysis of the motion of a conic curve. An image motion of a conic is assiciated with a group of invisible motions that do not change the character of the conic. BibRef 9302

Kanatani, K.[Ken_Ichi], Liu, W.[Wu],
3D Interpretation of Conics and Orthogonality,
CVGIP(58), No. 3, November 1993, pp. 286-301.
WWW Link. Geometry of a conic of a known shape from its projection. BibRef 9311

Kanatani, K.I., and Chou, T.C.,
Tracing Finite Motions Without Correspondence,
DARPA87(704-718). Given one object in the image the changing shape gives the motion parameters. BibRef 8700

Kanatani, K.I., (Gunma U. Japan, Now at UMd)
Tracing Planar Surface Motion from a Projection without Knowing the Correspondence,
CVGIP(29), No. 1, January 1985, pp. 1-12.
WWW Link. BibRef 8501
And:
Structure from Motion Without Correspondence: General Principle,
DARPA85(107-116). Motion is determined by the motion of the entire contour -- limited to single objects with well defined contours. The contour motion is given by the motion of the diameters. See also Structure and Motion from Optical Flow under Orthographic Projection. BibRef

Kanatani, K.I., (Gunma U. Japan)
Detecting the Motion of a Planar Surface by Line and Surface Integrals,
CVGIP(29), No. 1, January 1985, pp. 13-22.
WWW Link. And the above paper. BibRef 8501

Aloimonos, Y., and Herve, J.Y.,
Correspondenceless Stereo and Motion: Planar Surfaces,
PAMI(12), No. 5, May 1990, pp. 504-510.
IEEE DOI Rigid motion of a planar patch. BibRef 9005

Basu, A., and Aloimonos, Y.,
A Robust, Correspondenceless, Translation-Determining Algorithm,
IJRR(9), No. 5, 1990, pp. 35-59. BibRef 9000

Aloimonos, Y., and Basu, A.,
Shape and 3-D Motion from Contour without Point to Point Correspondences: General Principles,
CVPR86(518-527). Uses the global contour changes to compute the motion parameters. BibRef 8600

Ito, E., Aloimonos, Y.,
Is Correspondence Necessary for the Perception of Structure from Motion?,
DARPA88(921-929). BibRef 8800

Brown, C.M., Aloimonos, Y., Swain, M.J., Chou, P., Basu, A.,
Texture, Contour, Shape, and Motion,
PRL(5), 1987, pp. 151-168. BibRef 8700

Aloimonos, Y.[Yiannis (John)], Basu, A.[Anup], and Brown, C.M.,
Contour Orientation and Motion,
DARPA85(129-136). Find the changes in 2 contours by the change in shape. BibRef 8500

Aloimonos, Y., Brown, C.M.,
The Relationship Between Optical Flow and Surface Orientation,
ICPR84(542-545). BibRef 8400

Aloimonos, Y., and Rigoutsos, I.[Isidore],
Determining the 3-D Motion of a Rigid Planar Patch Without Correspondence, under Perspective Projection,
Motion86(167-174). BibRef 8600
And:
Determining the 3-D Motion of a Rigid Planar Patch Without Correspondence, under Perspective Projection, I. Planar Surfaces, II. Curved Surfaces,
AAAI-86(681-688). First: correspondence is assumed (at the gross - region - level). Second: merely accumulating the measurements for all the feature points in the two views. This is not really anything different. He argues that there cannot exist a robust algorithm for 3-D motion for a single camera, but has no proof, yet. BibRef

Korsten, M.J., and Houkes, Z.,
The Estimation of Geometry and Motion of a Surface from Image Sequences by Means of Linearization of a Parametric Model,
CVGIP(50), No. 1, April 1990, pp. 1-28.
WWW Link. Lots of theory. BibRef 9004

Lin, C.T.D.[Chih-Tzay D.], Goldgof, D.B.[Dmitry B], Huang, W.C.[Wen-Chen],
Motion Estimation from Scaled Orthographic Projections without Correspondences,
IVC(12), No. 2, March 1994, pp. 95-108.
WWW Link. BibRef 9403

Vieville, T., Lingrand, D., Gaspard, F.,
Implementing a Multi-Model Estimation Method,
IJCV(44), No. 1, August 2001, pp. 41-64.
DOI Link Revisit Kanatani's estimation method. See also Tracing Planar Surface Motion from a Projection without Knowing the Correspondence. 0108
BibRef


Bodis-Szomoru, A.[Andras], Riemenschneider, H.[Hayko], Van Gool, L.J.[Luc J.],
Fast, Approximate Piecewise-Planar Modeling Based on Sparse Structure-from-Motion and Superpixels,
CVPR14(469-476)
IEEE DOI 1409
SfM BibRef

Zhou, Z.[Zihan], Jin, H.L.[Hai-Lin], Ma, Y.[Yi],
Robust plane-based structure from motion,
CVPR12(1482-1489).
IEEE DOI 1208
BibRef

Nie, Q.[Qiong], Bouchafa, S.[Samia], Merigot, A.[Alain],
Voting spaces cooperation for 3D plane detection from monocular image sequences,
IPTA12(135-140)
IEEE DOI 1503
driver information systems BibRef

Xu, C.H.[Chang-Hai], Kuipers, B.[Benjamin], Murarka, A.[Aniket],
3D pose estimation for planes,
3DRR09(673-680).
IEEE DOI 0910
Track planes and estimate pose. BibRef

Dornaika, F.[Fadi], Davoine, F.[Franck],
SFM for Planar Scenes Using Image Derivatives,
ICIP05(III: 1088-1091).
IEEE DOI 0512
BibRef

Kamada, H., Yanagishta, H., Shiohara, M.,
Recognizing Moving Objects From Orthographic Views,
ICPR92(I:121-124).
IEEE DOI BibRef 9200

Hamano, T., Yasuno, T., Ishii, K.,
Direct Estimation of Structure from Non-Linear Motion by Voting Algorithm without Tracking and Matching,
ICPR92(I:505-508).
IEEE DOI BibRef 9200

Douterloigne, K.[Koen], Gautama, S.[Sidharta], Philips, W.[Wilfried],
Speeding Up Structure from Motion on Large Scenes Using Parallelizable Partitions,
ACIVS10(II: 13-21).
Springer DOI 1012
BibRef

Chapter on Motion Analysis -- Low-Level, Image Level Analysis, Mosaic Generation, Super Resolution, Shape from Motion continues in
Structure, Depth, and Shape from Motion .


Last update:Sep 18, 2017 at 11:34:11