18.9.8 Integration over a Sequence, Incremental Recovery

Chapter Contents (Back)
Sequences. Motion, Many Frames. Fusion.

Matthies, L.H.[Larry H.],
Dynamic Stereo Vision,
Ph.D.Thesis (CS). October 1989. BibRef 8910 CMU-CS-TR-89-195, CMU CS Dept. The long version of his work as reported in several other papers. BibRef

Matthies, L.H., Szeliski, R.S., and Kanade, T.,
Kalman Filter-based Algorithms for Estimating Depth from Image Sequences,
IJCV(3), No. 3. September 1989, pp. 209-238.
Springer DOI BibRef 8909
Earlier: DARPA88(199-213), also: BibRef CMU-RI-TR-88-1, CMU Robotics Institute. BibRef
And:
Incremental Estimation of Dense Depth Maps from Image Sequences,
CVPR88(366-374).
IEEE DOI Similar to the Broida type work in some aspects, but it is really multi-frame stereo analysis rather than motion analysis. BibRef

Charnley, D.[Debra], Blissett, R.[Rod],
Surface Reconstruction from Outdoor Image Sequences,
IVC(7), No. 1, February 1989, pp. 10-16.
Elsevier DOI BibRef 8902

Harris, C.G., Pike, J.M.,
3D positional integration from image sequences,
IVC(6), No. 2, May 1988, pp. 87-90.
Elsevier DOI BibRef 8805
Earlier: Alvey87(233-236). The points are tracked through the sequence, and their 3D locations are accurately determined by use of Kalman filters. The egomotion of the camera is also determined. BibRef

Hildreth, E.C., and Grzywacz, N.M., Adelson, E.H., and Inada, V.K.,
The Perceptual Buildup of Three-Dimensional Structure from Motion,
PandP(48), No. 1, 1990, pp. 19-26. BibRef 9000
And: MIT AI Memo-1141, August 1989. BibRef

Grzywacz, N.M.[Norberto M.], and Hildreth, E.C.[Ellen C.],
The Incremental Rigidity Scheme for Recovering Structure from Motion: Position vs. Velocity Based Formulations,
MIT AI Memo-845, October 1985.
WWW Link. BibRef 8510

Hildreth, E.C., and Grzywacz, N.M.,
The Incremental Recovery of Structure from Motion: Position vs. Velocity Based Formulations,
Motion86(137-143). This and the Ullman invited talk covered the topic. MIT has now learned that long range methods do not have thesame noise problems as the short range methods. The rigid object assumption is used to limit how much the 3-D model of the object changes from one view to the next, but given no 3-D to begin with, the actual structure is determined. The examples were for simple polygons, and they did not reall show what the input looked like - more a presentation problem. BibRef 8600

Hildreth, E.C.[Ellen C.], Ando, H.[Hiroshi], Andersen, R.[Richard], and Treue, S.[Stefan],
Recovering Three-Dimensional Structure from Motion with Surface Reconstruction,
MIT AI Memo-1314, December 1991. BibRef 9112

Levine, M.D.[Martin D.], Noble, P.B.[Peter B.], Youssef, Y.M.[Youssry M.],
Understanding Blood Cell Motion,
CVGIP(21), No. 1, January 1983, pp. 58-84.
Elsevier DOI BibRef 8301
And:
A Rule-Based System for Characterizing Blood Cell Motion,
ISPDSA83(663-709). BibRef

Choi, K.J.[Kwang-Jin], Park, S.H.[Sang-Hyun], Ko, H.S.[Hyeong-Seok],
Processing Motion Capture Data to Achieve Positional Accuracy,
GMIP(61), No. 5, September 1999, pp. 260-273. BibRef 9909

Aguiar, P.M.Q.[Pedro M.Q.], Moura, J.M.F.[Jose M.F.],
Three-dimensional modeling from two-dimensional video,
IP(10), No. 10, October 2001, pp. 1541-1551.
IEEE DOI 0110
BibRef
Earlier:
Image Motion Estimation: Convergence and Error Analysis,
ICIP01(II: 937-940).
IEEE DOI 0108
BibRef
Earlier:
A Fast Algorithm for Rigid Structure from Image Sequences,
ICIP99(III:125-129).
IEEE DOI BibRef
Earlier:
Video representation via 3D shaped mosaics,
ICIP98(I: 823-827).
IEEE DOI 9810
BibRef

Yokoya, N., Shakunaga, T., and Kanbara, M.,
Passive Range Sensing Techniques: Depth from Images,
IEICE(E82-D), No. 3, 1999, pp. 523-533. BibRef 9900

Sato, T., Kanbara, M., Yokoya, N., Takemura, H.,
3-D modeling of an outdoor scene by multi-baseline stereo using a long sequence of images,
ICPR02(III: 581-584).
IEEE DOI 0211
BibRef

Goldberger, J.[Jacob],
Reconstructing camera projection matrices from multiple pairwise overlapping views,
CVIU(97), No. 3, March 2005, pp. 283-296.
Elsevier DOI 0412
Factorization requires all views, others triplets of views. Only pairs of views required. BibRef

Ji, H.[Hui], Fermuller, C.[Cornelia],
A 3D Shape Constraint on Video,
PAMI(28), No. 6, June 2006, pp. 1018-1023.
IEEE DOI 0605
BibRef
Earlier:
Integration of Motion Fields through Shape,
CVPR05(II: 663-669).
IEEE DOI 0507
BibRef
Earlier:
Bias in Shape Estimation,
ECCV04(Vol III: 405-416).
Springer DOI 0405
BibRef

Pugeault, N.[Nicolas], Kruger, N.[Norbert],
Temporal accumulation of oriented visual features,
JVCIR(22), No. 2, February 2011, pp. 153-163.
Elsevier DOI 1102
Object model building; Visual representation; Feature tracking; Temporal filtering; Unscented Kalman filtering; Edge features; Multiple hypotheses tracking; Structure from motion BibRef

Pugeault, N., Woergoetter, F., Krueger, N.,
Accumulated Visual Representation for Cognitive Vision,
BMVC08(xx-yy).
PDF File. 0809
BibRef

Miura, M.[Masato], Arai, J.[Jun], Sasaki, H.[Hisayuki], Okui, M.[Makoto], Okano, F.[Fumio], Yamazaki, J.[Junichi], Sobue, S.I.[Shin-Ichi],
Extracting 3D images from lunar orbiter Kaguya data,
SPIE(Newsroom), March 14, 2011
DOI Link 1103
Geometric analysis of movies captured by a single high-definition TV camera gives depth to 2D images from space. BibRef

Lhuillier, M.[Maxime],
Incremental Fusion of Structure-from-Motion and GPS Using Constrained Bundle Adjustments,
PAMI(34), No. 12, December 2012, pp. 2489-2495.
IEEE DOI 1210
BibRef
Earlier:
Fusion of GPS and Structure-from-Motion Using Constrained Bundle Adjustments,
CVPR11(3025-3032).
IEEE DOI 1106
See also Generic and Real-Time Structure from Motion Using Local Bundle Adjustment. BibRef

Litvinov, V.[Vadim], Lhuillier, M.[Maxime],
Incremental Solid Modeling from Sparse Structure-from-Motion Data with Improved Visual Artifacts Removal,
ICPR14(2745-2750)
IEEE DOI 1412
BibRef
Earlier:
Incremental Solid Modeling from Sparse and Omnidirectional Structure-from-Motion Data,
BMVC13(xx-yy).
DOI Link 1402
Cameras. BibRef

Abdel-Wahab, M.[Mohammed], Wenzel, K.[Konrad], Fritsch, D.[Dieter],
Automated and Accurate Orientation of Large Unordered Image Datasets for Close-Range Cultural Heritage Data Recording,
PFG(2012), No. 6, 2012, pp. 679-689.
WWW Link. 1302
BibRef
Earlier:
Efficient Reconstruction of Large Unordered Image Datasets for High Accuracy Photogrammetric Applications,
AnnalsPRS(I-3), No. 2012, pp. 1-6.
HTML Version. 1209
BibRef

Toldo, R.[Roberto], Gherardi, R.[Riccardo], Farenzena, M.[Michela], Fusiello, A.[Andrea],
Hierarchical structure-and-motion recovery from uncalibrated images,
CVIU(140), No. 1, 2015, pp. 127-143.
Elsevier DOI 1509
Structure and motion BibRef

Arrigoni, F.[Federica], Rossi, B.[Beatrice], Fusiello, A.[Andrea],
Global Registration of 3D Point Sets via LRS Decomposition,
ECCV16(IV: 489-504).
Springer DOI 1611
BibRef
Earlier:
Robust and Efficient Camera Motion Synchronization via Matrix Decomposition,
CIAP15(I:444-455).
Springer DOI 1511
BibRef

Malapelle, F.[Francesco], Fusiello, A.[Andrea], Rossi, B.[Beatrice], Fragneto, P.[Pasqualina],
A data-fusion approach to motion-stereo,
SP:IC(43), No. 1, 2016, pp. 42-53.
Elsevier DOI 1604
Motion-stereo BibRef

Arrigoni, F.[Federica], Rossi, B.[Beatrice], Malapelle, F., Fragneto, P., Fusiello, A.[Andrea],
Robust Global Motion Estimation with Matrix Completion,
CloseRange14(63-70).
DOI Link 1411
BibRef

Arrigoni, F., Magri, L., Rossi, B., Fragneto, P., Fusiello, A.[Andrea],
Robust Absolute Rotation Estimation via Low-Rank and Sparse Matrix Decomposition,
3DV14(491-498)
IEEE DOI 1503
Approximation methods BibRef

Gherardi, R.[Riccardo], Farenzena, M.[Michela], Fusiello, A.[Andrea],
Improving the efficiency of hierarchical structure-and-motion,
CVPR10(1594-1600).
IEEE DOI 1006
BibRef
Earlier: A2, A3, A1:
Structure-and-motion pipeline on a hierarchical cluster tree,
3DIM09(1489-1496).
IEEE DOI 0910
BibRef
And: A2, A3, A1:
SAMANTHA: Structure-and-Motion Pipeline on a Hierarchical Cluster Tree,
Online2010.
WWW Link. How to structure sequence for computation. BibRef


Abuzaina, A.[Anas], Nixon, M.S.[Mark S.], Carter, J.N.[John N.],
3D motion estimation by evidence gathering,
ICPR16(1756-1761)
IEEE DOI 1705
BibRef
And:
3D Moving Object Reconstruction by Temporal Accumulation,
ICPR14(2125-2130)
IEEE DOI 1412
Fasteners, Feature extraction, Geometry, Kinematics, Motion estimation, Object recognition, Three-dimensional, displays. Angular velocity BibRef

Masiero, A., Guarnieri, A., Vettore, A., Pirotti, F.,
An ISVD-based Euclidian structure from motion for smartphones,
CloseRange14(401-406).
DOI Link 1411
BibRef

Zarrouati, N.[Nadege], Aldea, E.[Emanuel], Rouchon, P.[Pierre],
Robust depth regularization explicitly constrained by camera motion,
ICPR12(3606-3609).
WWW Link. 1302
camera on known path. BibRef

Ke, T., Zhang, Z.X., Huang, S.,
The Scanning Photogrammetry,
ISPRS12(XXXIX-B5:345-349).
DOI Link 1209
from rotating the camera BibRef

Rana, M.[Mayank], Taylor, G.[Graham], Spiro, I.[Ian], Bregler, C.[Christoph],
3D skeletal reconstruction from low-resolution multi-view images,
HAU3D12(58-63).
IEEE DOI 1207
At a distance for the objects. BibRef

Klopschitz, M.[Manfred], Irschara, A.[Arnold], Reitmayr, G.[Gerhard], Schmalstieg, D.[Dieter],
Robust Incremental Structure from Motion,
3DPVT10(xx-yy).
WWW Link. 1005
BibRef

Sur, F.[Frederic],
Robust Matching in an Uncertain World,
ICPR10(2350-2353).
IEEE DOI 1008
BibRef

Bhat, S.K.K.[Srikrishna K.K.], Berger, M.O.[Marie-Odile], Sur, F.[Frederic],
Visual Words for 3D Reconstruction and Pose Computation,
3DIMPVT11(326-333).
IEEE DOI 1109
BibRef

Bhat, S.K.K.[Srikrishna K.K.], Berger, M.O.[Marie-Odile], Simon, G.[Gilles], Sur, F.[Frederic],
Transitive Closure Based Visual Words for Point Matching in Video Sequence,
ICPR10(3300-3303).
IEEE DOI 1008
BibRef

Yamaguchi, T.[Tatsuhisa], Nobuhara, S.[Shohei], Matsuyama, T.[Takashi],
Cell-based object tracking method for 3D shape reconstruction using multi-viewpoint active cameras,
VS09(1306-1313).
IEEE DOI 0910
Localized calibration. BibRef

Jiang, Z.H.[Zhu-Han],
Object Modelling in Videos via Multidimensional Features of Colours and Textures,
DICTA09(154-161).
IEEE DOI 0912
Model a tracked object by distinctive features. BibRef

Knoblauch, D.[Daniel], Kuester, F.[Falko],
Focused Volumetric Visual Hull with Color Extraction,
ISVC09(II: 208-217).
Springer DOI 0911
voxels from moving object. BibRef

Knoblauch, D.[Daniel], Hess-Flores, M.[Mauricio], Duchaineau, M.[Mark], Kuester, F.[Falko],
Factorization of Correspondence and Camera Error for Unconstrained Dense Correspondence Applications,
ISVC09(I: 720-729).
Springer DOI 0911
BibRef

Li, C.[Chuan], Zheng, J.J.[Jin-Jin], Dang, C.Y.[Chuang-Yin], Zhou, H.J.[Hong-Jun],
A Method of 3D Reconstruction from Image Sequence,
CISP09(1-5).
IEEE DOI 0910
BibRef

Croitoru, I., Bogolin, S.V., Leordeanu, M.[Marius],
Unsupervised Learning from Video to Detect Foreground Objects in Single Images,
ICCV17(4345-4353)
IEEE DOI 1802
computer vision, image segmentation, neural nets, object detection, unsupervised learning, video signal processing, Visualization BibRef

Leordeanu, M.[Marius], Collins, R.[Robert],
Unsupervised Learning of Object Features from Video Sequences,
CVPR05(I: 1142-1149).
IEEE DOI 0507
Single or multiple objects, change in pose, low res video. Related features tend to move the same. BibRef

Le, H.V.[Ha Vu],
A structure-from-motion method for 3-d reconstruction of moving objects from multiple-view image sequences,
ICIP04(III: 1955-1958).
IEEE DOI 0505
BibRef
And:
A Structure-from-Motion Method: Use of Motion in Three-Dimensional Reconstruction of Moving Objects from Multiple-View Image Sequences,
3DPVT04(341-347).
IEEE DOI 0412
BibRef

Nguyen, H.V., Hanajík, M.,
3-D scene reconstruction from image sequences,
CAIP95(182-189).
Springer DOI 9509
From line segments. BibRef

Weng, J., Cui, Y., Ahuja, N., and Singh, A.,
Integration of Transitory Image Sequences,
CVPR94(966-969).
IEEE DOI BibRef 9400

Kim, Y.C., and Price, K.E.,
Multiple Frame Analysis of Translation Dominant Motion,
DARPA90(339-347). BibRef 9000 USC Computer VisionSimple application of multiple frame techniques to extract structure. BibRef

Ferrie, F.P., and Levine, M.D.,
Integrating Information from Multiple Views,
CVWS87(117-122). BibRef 8700
Earlier:
Piecing Together the 3D Shape of Moving Objects: An Overview,
CVPR85(574-584). Generate description, transformations, then the model. Integration of data for 3-D representation. Precursor to his thesis which should be dated about May 1986. BibRef

Altman, E.J., Ahuja, N.,
A Dynamical Systems Approach to Integration in Stereo,
DARPA90(423-427). BibRef 9000

Heel, J.[Joachim],
Temporal Surface Reconstruction,
CVPR91(607-612).
IEEE DOI BibRef 9100
And: Longer: MIT AI-TR-1296, May 1991.
WWW Link. BibRef

Heel, J.,
Temporally Integrated Surface Reconstruction,
ICCV90(292-295).
IEEE DOI BibRef 9000
And:
Dynamic Motion Vision,
DARPA89(702-713). BibRef
And: SPIE(1192), 1990, pp. 758-769. Using a sequence, predict the depth, predict the motion, use each prediction to estimate the other and update, using Bayesian estimation theory and a Kalman filter. BibRef

Heel, J.[Joachim],
Direct dynamic motion vision,
CRA90(II: 1142-1147). BibRef 9000

Heel, J.[Joachim],
Direct Estimation of Structure and Motion from Multiple Frames,
MIT AI Memo-1190, March 1990.
WWW Link. BibRef 9003

Heel, J., Rao, S.,
Temporal Integration of Visual Surface Reconstruction,
DARPA90(376-382). BibRef 9000

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


Last update:Jun 23, 2018 at 14:58:54