16.3 Motion Using Stereo Pairs or Depth, Multiple Cameras -- Features

Chapter Contents (Back)
Motion and Depth. Stereo and Motion. Motion and Stereo. See also Target Tracking, Multiple Sensors, Multiple Cameras, Multi-Camera Tracking. See also Multi-Target Tracking with Multiple Sensors, Stereo, Depth, Range.

Young, G.S., and Chellappa, R.,
3-D Motion Estimation Using a Sequence of Noisy Stereo Images: Models, Estimation, and Uniqueness Results,
PAMI(12), No. 8, August 1990, pp. 735-759.
IEEE DOI BibRef 9008
Earlier: CVPR88(710-716).
IEEE DOI Closed form solutions using quaternions. BibRef

Young, G.S., and Chellappa, R.,
Statistical Analysis of Inherent Ambiguities in Recovering 3-D Motion from a Noisy Flow Field,
PAMI(14), No. 10, October 1992, pp. 995-1013.
IEEE DOI BibRef 9210
Earlier: ICPR90(I: 371-377).

Chen, H.H.,
Determining Motion and Depth from Binocular Orthographic Views,
CVGIP(54), No. 1, July 1991, pp. 47-55.
Elsevier DOI BibRef 9107
Motion And Depth From Binocular Orthographic Views,
IEEE DOI The Z value from the stereo estimate is discarded thus giving an Orthographic projection (X and Y only). See also Using Motion from Orthographic Views to Verify 3-D Point Matches. BibRef

Abdel-Mottaleb, M., Chellappa, R., and Rosenfeld, A.,
Binocular Motion Stereo Using MAP Estimation,
IEEE DOI From the FOE to displacement to a depth map for axial motion. BibRef 9300

Huang, T.S., and Blostein, S.D.,
Robust Algorithms for Motion Estimation Based on Two Sequential Stereo Image Pairs,
CVPR85(518-523). (Univ. of Illinois) Motion, Estimation Evaluation. Motion with stereo using simulated data. Error analysis for the expected range of values, 9 points in the views are needed for good results. With more authors and a new title, essentially the same thing is in BibRef 8500 Motion86(45-46). BibRef

Blostein, S.D., and Huang, T.S.,
Estimating 3-D Motion from Range Data,
CAIA84(246-250). BibRef 8400

Hong, Z., and Ahuja, N.,
Target Tracking and Cumulative Depth Map Generation from Binocular Image Sequences,
IAS93(xx-yy). BibRef 9300

Mitiche, A.[Amar], and Bouthemy, P.[Patrick],
Tracking Modelled Objects Using Binocular Images,
CVGIP(32), No. 3, December 1985, pp. 384-396.
Elsevier DOI BibRef 8512
Representation and Tracking of Point Structures Using Stereovision,
CVWS84(118-124). Motion, Structure. From INRS-Telecommunications, 3 place du Commerce, Ile-des-Soeurs, PQ H3E 1H6. Develops a lot of equations to derive the motion from a given match which comes from stereo pairs. They assume that the stereo match can be done in a "brute force" manner - try all possible and choose the best. The camera parameters then give 3-D positions that are used in the motion match, which are then used to derive the motion parameters. BibRef

Meyer, F.G., Bouthemy, P.,
Region-Based Tracking Using Affine Motion Models in Long Image Sequences,
CVGIP(60), No. 2, September 1994, pp. 119-140.
DOI Link BibRef 9409
Region-Based Tracking in an Image Sequence,
Springer DOI BibRef

Deriche, R., Faugeras, O.D.,
Tracking Line Segments,
IVC(8), No. 4, November 1990, pp. 261-270. BibRef 9011
Earlier: ECCV90(259-268).
Springer DOI BibRef

Zhang, Z.Y., and Faugeras, O.D.,
Tracking and Grouping 3D Line Segments,
IEEE DOI BibRef 9000

Bascle, B., Bouthemy, P., Deriche, R., Meyer, F.,
Tracking Complex Primitives in an Image Sequence,
IEEE DOI BibRef 9400

Giai-Checa, B., Bouthemy, P., Vieville, T.,
Segment-Based Detection of Moving Objects in a Sequence of Images,
IEEE DOI BibRef 9400

Mutch, K.M.,
Determining Object Translation Information Using Stereoscopic Motion,
PAMI(8), No. 6, November 1986, pp. 750-755. BibRef 8611
And: With: Heiny, L.C.,
Calculating Object Size from Stereoscopic Motion,
CVPR86(183-187). Matching is eliminated by using a spot on white background. Assume the measured dimension is parallel to the camera baseline, and there is translation. BibRef

Aggarwal, J.K., and Magee, M.J.,
Determining Motion Parameters Using Intensity Guided Range Sensing,
PR(19), No. 2, 1986, pp. 169-180.
Elsevier DOI BibRef 8600
Earlier: ICPR84(538-541). Library models are matched to get the rotation and translation parameters of the object centers which provide the motion parameters. The other aspect is the intensity and range data combination work. See also Experiments in Intensity Guided Range Sensing Recognition of Three-Dimensional Objects. BibRef

Kim, Y.C., and Aggarwal, J.K.,
Determining Object Motion in a Sequence of Stereo Images,
RA(3), No. 5, December 1987, pp. 599-614. See also Positioning Three-Dimensional Objects Using Stereo Images. BibRef 8712

Zhang, Z.Y., and Faugeras, O.D.,
3D Dynamic Scene Analysis: A Stereo Based Approach,
SpringerBerlin, Heidelberg, 1992. BibRef 9200 BookSeveral proposed methods, both long and short sequences. Stereo results are superior to monocular. BibRef

Zhang, Z.Y., and Faugeras, O.D.,
Three-Dimensional Motion Computation and Object Segmentation in a Long Sequence of Stereo Frames,
IJCV(7), No. 3, April 1992, pp. 211-241.
Springer DOI Track 3-D components and estimate their motion using an extended Kalman filter. Then group tokens into objects based on similar motions. See also Motion of an Uncalibrated Stereo Rig: Self-Calibration and Metric Reconstruction. BibRef 9204

Zhang, Z.Y., Faugeras, O.D., and Ayache, N.J.,
Analysis of a Sequence of Stereo Scenes Containing Multiple Moving Objects Using Rigidity Constraints,
IEEE DOI Use the trinocular stereo system to generate depth images, match these and mark as moving anything that is not consistent. See also Trinocular Stereo Vision for Robotics. BibRef 8800

Zhang, Z.Y., and Faugeras, O.D.,
Estimation of Displacements from Two 3-D Frames Obtained from Stereo,
PAMI(14), No. 12, December 1992, pp. 1141-1156.
IEEE DOI Motion, Lines. Estimate displacement from two stereo frames using lines. It misses some important multi-frame motion and structure papers. There is a long bibliography even with these missing papers. BibRef 9212

Zhang, Z.Y., and Faugeras, O.D.,
Determining Motion from 3D Line Segment Matches: A Comparative Study,
IVC(9), No. 1, February 1991, pp. 10-19.
Elsevier DOI BibRef 9102
Earlier: BMVC90(xx-yy).
PDF File. 9009
SVD. Since the 3D data is extracted, it is noisy. Compares EKF, general minimization, and SVD. BibRef

Zhang, Z.Y.[Zheng-You], Faugeras, O.D.[Olivier D.],
Finding Planes and Clusters of Objects from 3D Line Segments with Application to 3D Motion Determination,
CVGIP(60), No. 3, November 1994, pp. 267-284.
DOI Link BibRef 9411
Finding clusters and planes from 3D line segments with application to 3D motion determination,
Springer DOI 9205
Find planes based on clusters of line segments BibRef

Zhang, Z.Y.[Zheng-You],
Motion and Structure of Four Points from One Motion of a Stereo Rig with Unknown Extrinsic Parameters,
PAMI(17), No. 12, December 1995, pp. 1222-1227.
IEEE DOI BibRef 9512
Motion of a Stereo Rig: Strong, Weak, and Self-Calibration,
ACCV95(1274-1281). BibRef
Earlier: CVPR93(556-561).
IEEE DOI Two stereo pairs, four points, determine camera motion, camera positions, and structure. See also Motion of an Uncalibrated Stereo Rig: Self-Calibration and Metric Reconstruction. BibRef

Zhang, Z.Y.[Zheng-You],
An automatic and robust algorithm for determining motion and structure from two perspective images,
Springer DOI 9509

Faugeras, O.D., Ayache, N., Zhang, Z.Y.,
A Preliminary Investigation of the Problem of Determining Ego- and Object Motions from Stereo,
ICPR88(I: 242-246).
IEEE DOI BibRef 8800

Tsukiyama, T.[Toshifumi], Huang, T.S.[Thomas S.],
Motion Stereo for Navigation of Autonomous Vehicles in Man-Made Environments,
PR(20), No. 1, 1987, pp. 105-113.
Elsevier DOI BibRef 8700
Earlier: ICPR86(165-168). BibRef
Motion Stereo for Navigation of Autonomous Vehicles in a Passageway,
CVWS85(148-155). BibRef

Tsukiyama, T., Shirai, Y.,
Detection of the Movements of Men for Autonomous Vehicles,
IJCAI79(908-910). BibRef 7900

Shieh, J.Y., Zhuang, H., Sudhakar, R.,
Motion Estimation From A Sequence Of Stereo Images: A Direct Method,
SMC(24), No. 7, July 1994, pp. 1044-1053. BibRef 9407

Liao, W.H.[Wen-Hung], Aggarwal, S.J., Aggarwal, J.K.,
The Reconstruction of Dynamic 3D Structure of Biological Objects Using Stereo Microscope Images,
MVA(9), No. 4, 1997, pp. 166-178.
Springer DOI BibRef 9700
Reconstruction of dynamic 3-D structures of biological objects using stereo microscopy,
ICIP94(III: 731-735).
Nonrigid Motion. Image registration by correlation, region of interest using motion based segmentation, stereo and motion BibRef

Shih, S.W., Hung, Y.P., Lin, W.S.,
New Closed-Form Solution for Kinematic Parameter-Identification of a Binocular Head Using Point Measurements,
SMC-B(28), No. 2, April 1998, pp. 258-267.
IEEE Top Reference. 9804

Liao, W.H., Aggarwal, J.K.,
Cooperative Matching Paradigm for the Analysis of Stereo Image Sequences,
IJIST(9), No. 4, 1998, pp. 192-200. 9808

Ho, P.K.[Pui-Kuen], Chung, R.[Ronald],
Stereo-Motion with Stereo and Motion in Complement,
PAMI(22), No. 2, February 2000, pp. 215-220.
Stereo-Motion That Complements Stereo and Motion Analyses,
Decompose 3D, into motions, stereogeometry. BibRef

Dornaika, F.[Fadi], Chung, C.R.,
Stereo geometry from 3D ego-motion streams,
SMC-B(33), No. 2, April 2003, pp. 308-323.
IEEE Abstract. 0308

Dornaika, F.[Fadi], Chung, R.[Ronald],
Cooperative Stereo-Motion: Matching and Reconstruction,
CVIU(79), No. 3, September 2000, pp. 408-427.
DOI Link 0008
Stereo Correspondence from Motion Correspondence,
CVPR99(I: 70-75).

Ku, J.S.[Ja Seong], Lee, K.M.[Kyoung Mu], Lee, S.U.[Sang Uk],
Multi-image matching for a general motion stereo camera model,
PR(34), No. 9, September 2001, pp. 1701-1712.
Elsevier DOI 0108
Earlier: ICIP98(II: 608-612).

Heo, Y.S.[Yong Seok], Lee, K.M.[Kyoung Mu], Lee, S.U.[Sang Uk],
Illumination and camera invariant stereo matching,

Qian, G.[Gang], Chellappa, R.[Rama], Zheng, Q.F.[Qin-Fen],
Robust structure from motion estimation using inertial data,
JOSA-A(18), No. 12, December 2001, pp. 2982-2997.
WWW Link. 0201
Earlier: A1, A3, A2:
Reduction of Inherent Ambiguities in Structure from Motion Problem Using Inertial Data,
ICIP00(Vol I: 204-207).

Qian, G.[Gang], Chellappa, R., Zheng, Q.F.[Qin-Fen],
Bayesian structure from motion using inertial information,
ICIP02(III: 425-428).
Use inertial guidance info to help in the SfM solution. BibRef

Qian, G., Kale, A., Chellappa, R.,
Robust Estimation of Motion and Structure Using a Discrete h8 Filter,
ICIP00(Vol III: 616-619).

Qian, G., Chellappa, R., Zheng, Q.F., Ortolf, J.,
Camera Motion Estimation Using Monocular Image Sequences and Inertial Data,
UMD--TR3997, March 1999.
WWW Link. BibRef 9903

Qian, G.[Gang], Chellappa, R., Zheng, Q.F.[Qin-Fen],
Bayesian Algorithms for Simultaneous Structure From Motion Estimation of Multiple Independently Moving Objects,
IP(14), No. 1, January 2005, pp. 94-109.
Robust bayesian cameras motion estimation using random sampling,
ICIP04(II: 1361-1364).
A bayesian approach to simultaneous motion estimation of multiple independently moving objects,
ICPR02(III: 309-314).

Qian, G.[Gang], Chellappa, R.[Rama],
Structure from Motion Using Sequential Monte Carlo Methods,
IJCV(59), No. 1, August 2004, pp. 5-31.
DOI Link 0404
Earlier: ICCV01(II: 614-621).
Random sampling. BibRef

Kaminski, J.Y.[Jeremy Yirmeyahu], Teicher, M.[Mina],
A General Framework for Trajectory Triangulation,
JMIV(21), No. 1, July 2004, pp. 27-41.
DOI Link 0409
General Trajectory Triangulation,
ECCV02(II: 823 ff.).
Springer DOI 0205
Stereo with motion and non-synchronized cameras. Use the trajectory. BibRef

Kim, J.S.[Jun-Sik], Hwangbo, M.[Myung], Kanade, T.[Takeo],
Spherical approximation for multiple cameras in motion estimation: Its applicability and advantages,
CVIU(114), No. 10, October 2010, pp. 1068-1083.
Elsevier DOI 1003
Camera motion estimation; Multiple cameras; Spherical approximation; Camera calibration; Structure from motion BibRef

Chandraker, M.[Manmohan], Lim, J.W.[Jong-Woo], Kriegman, D.[David],
Moving in stereo: Efficient structure and motion using lines,

Zhang, Q.L.[Qi-Long], Pless, R.,
Fusing video and sparse depth data in structure from motion,
ICIP04(V: 3403-3406).

Wan, A.S.K., Siu, A.M.K., Lau, R.W.H., Ngo, C.W.,
A robust method for recovering geometric proxy from multiple panoramic images,
ICIP04(II: 1369-1372).
3D motion from wide baseline cameras with noisy matches. BibRef

Demirdjian, D.[David], Horaud, R.[Radu],
A Projective Framework for Scene Segmentation in the Presence of Moving Objects,
CVPR99(I: 2-8).
IEEE DOI Given a sequence of pairs, and the corresponding points, this is what you can do. BibRef 9900

Sparr, G., Lindström, P.,
Euclidean Reconstruction and Calibration from Known Placements of Uncalibrated and Uncorrelated Cameras,
SCIA99(Computer Vision). BibRef 9900

Sparr, G.[Gunnar],
Euclidean and Affine Structure/Motion for Uncalibrated Cameras from Affine Shape and Subsidiary Information,
SMILE98(xx-yy). BibRef 9800
Simultanious Reconstruction of Scene Structure and Camera Locations from Uncalibrated Image Sequences,
ICPR96(I: 328-333).
A Common Framework for Kinetic Depth, Reconstruction and Motion for Deformable Objects,
Springer DOI (Lund Univ./LTH, S) BibRef

Weinshall, D.[Daphna], Anandan, P., Irani, M.[Michal],
From Ordinal to Euclidean Reconstruction with Partial Scene Calibration,
SMILE98(xx-yy). BibRef 9800

Navab, N., Deriche, R., and Faugeras, O.D.,
Recovering 3D Motion and Structure from Stereo and 2D Token Tracking Cooperation,
IEEE DOI Stereo and optical flow, using lines. BibRef 9000

Waldmann, J., Merhav, S.,
Fusion of Stereo and Motion Vision for 3-D Reconstruction,
IEEE DOI BibRef 9200

Weng, J., Huang, T.S.,
Complete Structure and Motion from Two Monocular Sequences without Stereo Correspondence,
IEEE DOI BibRef 9200

Gambotto, J.P.,
Determining Stereo Correspondences and Egomotion from a Sequence of Stereo Images,
ICPR90(I: 259-262).
IEEE DOI Trinocular view, motion helps stereo helps motion. BibRef 9000

Thacker, N.A., Zheng, Y., and Blackbourn, R.,
Using a Combined Stereo/Temporal Matcher to Determine Ego-motion,
PDF File. Matches based on second derivatives at corners. BibRef 9000

Asada, M.[Minoru], Tsuji, S.[Saburo],
Inferring Motion of Cylindrical Object from Shape Information,
IJCAI83(1032-1034). BibRef 8300
Inferring Motion of Cylindrical Object from Shading,
CVPR83(240-245). Shading information is used to segment the scene (synthetic images are used). The motion can be derived by matching. ( See also Automatic Analysis of Moving Images. ) BibRef

Tsuji, S., Morizono, A., Kuroda, S.,
Understanding a Simple Cartoon Film by a Computer Vision System,
IJCAI77(609-610). BibRef 7700

Chapter on Motion -- Feature-Based, Long Range, Motion and Structure Estimates, Tracking, Surveillance, Activities continues in
Motion Using Depth Information .

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