Merging multiple views using a spherical representation
08 Feb 1994-pp 124-131
TL;DR: A new method for building a 3-D model from a set of range images based on matching the spherical representations of an object between the views, which can merge data of free-form surfaces obtained from arbitrary viewing directions, with no prior knowledge of the poses.
Abstract: This paper proposes a new method for building a 3-D model from a set of range images. The method can merge data of free-form surfaces obtained from arbitrary viewing directions, with no prior knowledge of the poses. Our approach is based on matching the spherical representations of an object between the views. To obtain the spherical representation, we deform a discrete mesh to fit the object surface. A variation of the Gaussian curvature metric, which we call simplex angle, is computed at each node on the deformed mesh and mapped to a coordinate on the unit sphere. The transformation of the objects is computed by comparing the simplex angle measure at each node on the unit sphere. The transformation which produces the minimum errors is selected as the best match. We have implemented this method, applied the method to range images of objects from arbitrary viewpoints, and demonstrated the applicability for modeling from observation. >
Citations
More filters
01 May 2000-Graphical Models \/graphical Models and Image Processing \/computer Vision, Graphics, and Image Processing
TL;DR: This paper addresses the problem of scanning both the color and geometry of real objects and displaying realistic images of the scanned objects from arbitrary viewpoints with a complete system that uses a stereo camera setup with active lighting to scan the object surface geometry and color.
Abstract: This paper addresses the problem of scanning both the color and geometry of real objects and displaying realistic images of the scanned objects from arbitrary viewpoints. We describe a complete system that uses a stereo camera setup with active lighting to scan the object surface geometry and color. Scans expressed in sensor coordinates are registered into a single object-centered coordinate system by aligning both the color and geometry where the scans overlap. The range data are integrated into a surface model using a robust hierarchical space carving method. The fit of the resulting approximate mesh to data is improved and the mesh structure is simplified using mesh optimization methods. In addition, a method for view-dependent texturing of the reconstructed surfaces is described. The method projects the color data from the input images onto the surface model and blends the various images depending on the location of the viewpoint and other factors such as surface orientation.
99 citations
Cites methods from "Merging multiple views using a sphe..."
...[23] used a representation that decouples rotation and translation and noted that solving the translation is easy once the relative rotation has been solved....
[...]
TL;DR: The object recognition system named RIO (relational indexing of objects), which contains a number of new techniques, is able to recognize 3D objects having planar, cylindrical, and threaded surfaces in complex, multiobject scenes.
74 citations
Cites background from "Merging multiple views using a sphe..."
...Most systems fall into three main categories: (1) systems that use intensity data alone [1, 7, 9, 11, 35, 42, 43, 48, 59], (2) systems that use range data alone [5, 10, 24, 26, 27, 38, 30, 37, 40, 50], and (3) systems that use both range and intensity (sometimes including color) data [31, 34, 52]....
[...]
25 Oct 1996
TL;DR: Novel algorithms to automatically construct object-localization models from many images of the object are presented, and a consensus-search approach to determine which parts of the image justifiably constitute inclusion in the model is presented.
Abstract: : Being able to accurately estimate an object's pose (location) in an image is important for practical implementations and applications of object recognition. Recognition algorithms often trade off accuracy of the pose estimate for efficiency -- usually resulting in brittle and inaccurate recognition. One solution is object localization -- a local search for the object's true pose given a rough initial estimate of the pose. Localization is made difficult by the unfavorable characteristics (for example, noise, clutter, occlusion and missing data) of real images. In this thesis, we present novel algorithms for localizing 3D objects in 3D range-image data (3D-3D localization) and for localizing 3D objects in 2D intensity-image data (3D-2D localization). Our localization algorithms utilize robust statistical techniques to reduce the sensitivity of the algorithms to the noise, clutter, missing data, and occlusion which are common in real images. Our localization results demonstrate that our algorithms can accurately determine the pose in noisy, cluttered images despite significant errors in the initial pose estimate. Acquiring accurate object models that facilitate localization is also of great practical importance for object recognition. In the past, models for recognition and localization were typically created by hand using computer-aided design (CAD) tools. Manual modeling suffers from expense and accuracy limitations. In this thesis, we present novel algorithms to automatically construct object-localization models from many images of the object. We present a consensus-search approach to determine which parts of the image justifiably constitute inclusion in the model. Using this approach, our modeling algorithms are relatively insensitive to the imperfections and noise typical of real image data. Our results demonstrate that our modeling algorithms can construct very accurate geometric models from rather noisy input data.
70 citations
Cites background from "Merging multiple views using a sphe..."
...Thesis Committee: Katsushi Ikeuchi, Chair Martial Hebert Steven Shafer, CMU/Microsoft Eric Grimson, MIT c 1996 Mark D. Wheeler This research has been supported in part by the Advanced Research Projects Agency under the Department of the Army, Army Research Office grant number DAAH04-94-G-0006, and in part by the Department of the Navy, Office of Naval Research grant numbers N00014-95-1-0591 and N00014-93-1-1220....
[...]
...I would first like to thank Katsushi Ikeuchi, my advisor and friend for much of my 7 year stay at Carnegie Mellon....
[...]
...Higuchi, Delingette, Hebert and Ikeuchi [63] presented the simplex angle image (SAI) for representing 3D objects as attributes (simplex angles which are related to surface curvature) spread among the nodes of a tessellated a sphere....
[...]
TL;DR: In this article, the curvature method of Bash and Ullman is used to model points on the object rim, while stereo matching is used for internal edge points to predict an object edge map from pose parameters.
Abstract: A method is presented for computing the pose of rigid 3D objects with arbitrary curved surfaces. Given an input image and a candidate object model and aspect, the method will verify whether or not the object is present and if so, report pose parameters. The curvature method of Bash and Ullman is used to model points on the object rim, while stereo matching is used for internal edge points. The model allows an object edge-map to be predicted from pose parameters. Pose is computed via an iterative search for the best pose parameters. Heuristics are used so that matching can succeed in the presence of occlusion and artifact and without resetting to use of corresponding salient feature points. Bench tests and simulations show that the method almost always converges to ground truth pose parameters for a variety of objects and for a broad set of starting parameters in the same aspect.
35 citations
[...]
TL;DR: This paper has introduced a new topological organization called modeling wave set (MWS) where an n-connectivity relationship is established where an object is simultaneously modeled in n subspaces of features, corresponding to n different viewing directions of the object.
25 citations
References
More filters
TL;DR: In this paper, the authors describe a general-purpose representation-independent method for the accurate and computationally efficient registration of 3D shapes including free-form curves and surfaces, based on the iterative closest point (ICP) algorithm, which requires only a procedure to find the closest point on a geometric entity to a given point.
Abstract: The authors describe a general-purpose, representation-independent method for the accurate and computationally efficient registration of 3-D shapes including free-form curves and surfaces. The method handles the full six degrees of freedom and is based on the iterative closest point (ICP) algorithm, which requires only a procedure to find the closest point on a geometric entity to a given point. The ICP algorithm always converges monotonically to the nearest local minimum of a mean-square distance metric, and the rate of convergence is rapid during the first few iterations. Therefore, given an adequate set of initial rotations and translations for a particular class of objects with a certain level of 'shape complexity', one can globally minimize the mean-square distance metric over all six degrees of freedom by testing each initial registration. One important application of this method is to register sensed data from unfixtured rigid objects with an ideal geometric model, prior to shape inspection. Experimental results show the capabilities of the registration algorithm on point sets, curves, and surfaces. >
17,598 citations
TL;DR: A new approach is proposed which works on range data directly and registers successive views with enough overlapping area to get an accurate transformation between views and is performed by minimizing a functional which does not require point-to-point matches.
2,850 citations
09 Apr 1991
TL;DR: The authors propose an approach that works on range data directly and registers successive views with enough overlapping area to get an accurate transformation between views and performs a functional that does not require point-to-point matches.
Abstract: The problem of creating a complete model of a physical object is studied. Although this may be possible using intensity images, the authors use range images which directly provide access to three-dimensional information. The first problem that needs to be solved is to find the transformation between the different views. Previous approaches have either assumed this transformation to be known (which is extremely difficult for a complete model) or computed it with feature matching (which is not accurate enough for integration. The authors propose an approach that works on range data directly and registers successive views with enough overlapping area to get an accurate transformation between views. This is performed by minimizing a functional that does not require point-to-point matches. Details are given of the registration method and modeling procedure, and they are illustrated on range images of complex objects. >
2,157 citations
01 Jan 1992
TL;DR: In this article, a least-squares technique is used to estimate 3D motion from the point correspondences, which reduces the average distance between curves in two sets, and yields an accurate motion estimate.
Abstract: Geometric matching in general is a difficult unsolved problem in computer vision. Fortunately, in many pratical applications, some a priori knowledge exists which considerably simplifies the problem. In visual navigation, for example, the motion between successive positions is usually either small or approximately known, but a more precise registration is required for environment modeling. The algorithm described in this report meets this need. Objects are represented by free-form curves, i.e., arbitrary spaces curves of the type found in practice. A curve is available in the form of a set of chained points. The proposed algorithm is based on iteratively matching points on one curve to the closest points on the other. A least-squares technique is used to estimate 3-D motion from the point correspondences, which reduces the average distance between curves in two sets. Both synthetic and real data have been used to test the algorithm, and the results show that it is efficient and robust, and yields an accurate motion estimate. The algorithm can be easily extended to solve similar problems such as 2-D curve matching and 3-D surface matching.
1,986 citations
TL;DR: The ``volume segment'' representation presented in this paper is a volumetric representation that facilitates modification yet is descriptive of surface detail in a bounding volume approximating the object generating the contours.
Abstract: Occluding contours from an image sequence with view-point specifications determine a bounding volume approximating the object generating the contours. The initial creation and continual refinement of the approximation requires a volumetric representation that facilitates modification yet is descriptive of surface detail. The ``volume segment'' representation presented in this paper is one such representation.
500 citations