Extended structure and motion analysis from monocular image sequences

Cui, Ning; Weng, Juyang; Cohen, Paul R.

doi:10.1109/iccv.1990.139523

Cited by 30 publications

(25 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent research focuses on extraction of shape and motion from longer image sequences [Kumar et al, 1989;Debrunner and Ahuja, 1990;Cui et al, 1990;Tomasi and Kanade, 1990;Tomasi and Kanade, 1991;Chen and Tsuji, 1992]. Debrunner and Ahuja [1990] provide closedform expressions for shape and motion assuming that motion is constant over the sequence (see also [Broida and Chellappa, 1991;Kumar et al, 1989;Weng et al, 1993]).…”

Section: Previous Workmentioning

confidence: 99%

“…Incremental solutions for multiple motions are computed by taking advantage of the redundancy of measurements. Cui, Weng, and Cohen [1990] use an optimal estimation technique (non-linear least squares) between each pair of frames, and an extended Kalman filter to accumulate information over time. Tomasi and Kanade [1991] use a factorization method which extracts shape and motion from an image stream without computing camera-centered depth.…”

Section: Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Recovering 3D Shape and Motion from Image Streams Using Nonlinear Least Squares

Szeliski

Kang

1994

Journal of Visual Communication and Image Representation

235

View full text Add to dashboard Cite

The Cambridge laboratory became operational in 1988 and is located at One Kendall Square, near MIT. CRL engages in computing research to extend the state of the computing art in areas likely to be important to Digital and its customers in future years. CRL's main focus is applications technology; that is, the creation of knowledge and tools useful for the preparation of important classes of applications. CRL AbstractThe simultaneous recovery of 3D shape and motion from image sequences is one of the more difficult problems in computer vision. Classical approaches to the problem rely on using algebraic techniques to solve for these unknowns given two or more images. More recently, a batch analysis of image streams (the temporal tracks of distinguishable image features) under orthography has resulted in highly accurate reconstructions. We generalize this approach to perspective projection and partial or uncertain tracks by using a non-linear least squares technique. While our approach requires iteration, it quickly converges to the desired solution, even in the absence of a priori knowledge about the shape or motion. Important features of the algorithm include its ability to handle partial point tracks, to use line segment matches and point matches simultaneously, and to use an object-centered representation for faster and more accurate structure and motion recovery. We also show how a projective (as opposed to scaled rigid) structure can be recovered when the camera calibration parameters are unknown.

show abstract

Section: Previous Workmentioning

confidence: 99%

Section: Previous Workmentioning

confidence: 99%

Recovering 3D Shape and Motion from Image Streams Using Nonlinear Least Squares

Szeliski

Kang

1994

Journal of Visual Communication and Image Representation

235

View full text Add to dashboard Cite

show abstract

“…Most works deal with non-transitory image sequences, and successful improvements have been achieved in their fusion (e.g., [4], [6], [8], [1]). Experiments for scene construction from transitory image sequence only started recently, and we have so far seen two efforts by Cui et al [2] and Tomasi and Kanade [9], respectively. In Cui et al [2], some relative accuracy was reported from a transitory image sequence, which indicated that the accuracy was not further reduced once incoming and exiting feature points are comparable.…”

Section: Introductionmentioning

confidence: 99%

Transitory image sequences, asymptotic properties, and estimation of motion and structure

Weng

Cui

Ahuja

1997

IEEE Trans. Pattern Anal. Machine Intell.

Self Cite

View full text Add to dashboard Cite

Abstract-A transitory image sequence is one in which no scene element is visible through the entire sequence. When a camera system scans a scene which cannot be covered by a single view, the image sequence is transitory. This article deals with some major theoretical and algorithmic issues associated with the task of estimating structure and motion from transitory image sequences. It is shown that integration with a transitory sequence has properties that are very different from those with a nontransitory one. Two representations, world-centered (WC) and camera-centered (CC), behave very differently with a transitory sequence. The asymptotic error rates derived in this article indicate that one representation is significantly superior to the other, depending on whether one needs camera-centered or world-centered estimates. To establish the tightness of these error rates, it has been shown that these reachable error rates are in fact the lowest possible given by a theoretical lower error bound, the Cramér-Rao error bound. Based on these results, we introduce an efficient "cross-frame" estimation technique for the CC representation. For the WC representation, our analysis indicates that a good technique should be based on camera global pose instead of interframe motions. In addition to testing with synthetic data, rigorous experiments were conducted with real-image sequences taken by a fully calibrated camera system. The comparison of the experimental results with the ground truth has demonstrated that a good accuracy can be obtained from transitory image sequences.

show abstract

“…For monocular sequences, we have the problem of unknown scale for the estimated structure [2]. The scale factor of any two consecutive images depends on the scale factor of the first two images.…”

Section: )mentioning

confidence: 99%

Integrated 3D analysis of flight image sequences

Sull

Ahuja

1994

Computer Vision — ECCV '94

View full text Add to dashboard Cite

Abstract. This paper is concerned with three-dimensional (3D) analysis of images showing 3D motion of an observer relative to a scene. It presents an approach to recovering 3D motion and structure parameters from multiple features present in a monocular image sequence such as points, regions, lines, texture gradient and vanishing line. For concreteness, the paper focuses on flight images of a planar, textured surface. In this paper, a linear integrated estimation method using two views is developed. Then, for robust estimation, a nonlinear integrated estimation method using multiple frames is presented. The integration of information in these diverse features is carried out using minimization of image errors. To reduce computation, a sequential-batch method is used to compute motion and structure. Performance is evaluated through simulations and experiments with a real image sequence digitized from a commercially available laserdisc of films taken from flying aircrafts.

show abstract

Extended structure and motion analysis from monocular image sequences

Cited by 30 publications

References 12 publications

Recovering 3D Shape and Motion from Image Streams Using Nonlinear Least Squares

Recovering 3D Shape and Motion from Image Streams Using Nonlinear Least Squares

Transitory image sequences, asymptotic properties, and estimation of motion and structure

Integrated 3D analysis of flight image sequences

Contact Info

Product

Resources

About