How far 3D shapes can be understood from 2D silhouettes

Laurentini, Aldo

doi:10.1109/34.368170

Cited by 135 publications

(72 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As can be seen from the figure, estimated trajectories are quite accurate, especially for the first four objects. As shown by Figure 6, the objects' visual hulls [3,16] are also recovered quite well. In fact, most inaccuracies are not so much due to errors in the recovered projection matrices as to the fact that a limited set of camera positions was used to construct each model.…”

Section: Geometric Consistency Constraintsmentioning

confidence: 57%

“…The single-camera case is more difficult, and all approaches proposed so far have either been limited to circular motions [11,18,28], required a reasonable guess to bootstrap an iterative estimation process [2,6], or been limited to synthetic data [26]. Likewise, all published methods for computing visual hulls [16] from image silhouettes, dating back to Baumgart's 1974 thesis [3], have assumed that the camera configurations were known a priori.…”

Section: Introductionmentioning

confidence: 99%

“…This algorithm uses the signature representation of the dual of image silhouettes, proposed in [1] in the object recognition context, to identify promising correspondences, and it exploits the redundancy of multiple epipolar geometries [17] to retain the consistent ones. The visual hull [3,16] of the observed solid is finally reconstructed from the recovered viewpoints. We have implemented this algorithm, and tested it on six real image sequences.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Structure and Motion from Images of Smooth Textureless Objects

Furukawa

Sethi

Ponce

et al. 2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper addresses the problem of estimating the 3D shape of a smooth textureless solid from multiple images acquired under orthographic projection from unknown and unconstrained viewpoints. In this setting, the only reliable image features are the object's silhouettes, and the only true stereo correspondence between pairs of silhouettes are the frontier points where two viewing rays intersect in the tangent plane of the surface. An algorithm for identifying geometrically-consistent frontier points candidates while estimating the cameras' projection matrices is presented. This algorithm uses the signature representation of the dual of image silhouettes to identify promising correspondences, and it exploits the redundancy of multiple epipolar geometries to retain the consistent ones. The visual hull of the observed solid is finally reconstructed from the recovered viewpoints. The proposed approach has been implemented, and experiments with six real image sequences are presented, including a comparison between ground-truth and recovered camera configurations, and sample visual hulls computed by the algorithm.

show abstract

Section: Geometric Consistency Constraintsmentioning

confidence: 57%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Structure and Motion from Images of Smooth Textureless Objects

Furukawa

Sethi

Ponce

et al. 2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…On the other hand, if only sparse, discrete views are available and the object has relatively complex topologies, volume intersection techniques [38,39] can be employed to produce a volumetric model which represents the visual hull [40,41] of the object. A brief review and comparison of existing approaches for model reconstruction from apparent contours are given in the following subsections.…”

Section: Algorithm 1 Motion Estimation From Apparent Contoursmentioning

confidence: 99%

Structure and motion estimation from apparent contours under circular motion

Wong

Mendonça

Cipolla

2002

Image and Vision Computing

View full text Add to dashboard Cite

In this paper we address the problem of recovering structure and motion from the apparent contours of a smooth surface. Fixed image features under circular motion and their relationships with the intrinsic parameters of the camera are exploited to provide a simple parameterization of the fundamental matrix relating any pair of views in the sequence. Such a parameterization allows a trivial initialization of the motion parameters, which all bear physical meanings. It also greatly reduces the dimension of the search space for the optimization problem, which can now be solved using only 2 epipolar tangents. In contrast to previous methods, the motion estimation algorithm introduced here can cope with incomplete circular motion and more widely spaced images. Existing techniques for model reconstruction from apparent contours are then reviewed and compared. Experiment on real data has been carried out and the 3D model reconstructed from the estimated motion is presented.

show abstract

“…Some are based on the visual hull concept that requires a number of cameras from all directions [18,14,15]. Some try to match the silhouette segment directly [23], which is a very challenging task because image silhouettes, usually on the occlusion boundaries, may represent different parts of the object in the stereo images, i.e., the silhouette in the right image is not the same as in the left.…”

Section: Plane Approximationmentioning

confidence: 99%

Combining approximate geometry with view-dependent texture mapping - a hybrid approach to 3D video teleconferencing

Kurashima

Yang

Lastra

Proceedings. XV Brazilian Symposium on Computer Graphics and Image Processing

View full text Add to dashboard Cite

Abstract. We present a hybrid system using computer vision and graphics methods that effectively combines fast automatic 3D model extraction with view-dependent texture mapping for the purpose of real-time person-toperson 3D video teleconferencing. In our approach the problem of creating a dense disparity map is bypassed. We compute only a sparse geometric model, which is textured from multiple cameras using an image-based rendering approach. Using a commodity graphics card and three personal computers arranged in a cluster, we achieve a performance of 3 to 4 frames per second. We demonstrate the ability to provide a 3D tele-immersive environment with our system.

show abstract

How far 3D shapes can be understood from 2D silhouettes

Cited by 135 publications

References 14 publications

Structure and Motion from Images of Smooth Textureless Objects

Structure and Motion from Images of Smooth Textureless Objects

Structure and motion estimation from apparent contours under circular motion

Combining approximate geometry with view-dependent texture mapping - a hybrid approach to 3D video teleconferencing

Contact Info

Product

Resources

About