Showing papers by "Luc Van Gool published in 2000"

Wide Baseline Stereo Matching based on Local, Affinely Invariant Regions

Abstract: ‘Invariant regions’ are image patches that automatically deform with changing viewpoint as to keep on covering identical physical parts of a scene. Such regions are then described by a set of invariant features, which makes it relatively easy to match them between views and under changing illumination. In previous work, we have presented invariant regions that are based on a combination of corners and edges. The application discussed then was image database retrieval. Here, an alternative method for extracting (affinely) invariant regions is given, that does not depend on the presence of edges or corners in the image but is purely intensity-based. Also, we demonstrate the use of such regions for another application, which is wide baseline stereo matching. As a matter of fact, the goal is to build an opportunistic system that exploits several types of invariant regions as it sees fit. This yields more correspondences and a system that can deal with a wider range of images. To increase the robustness of the system even further, two semi-local constraints on combinations of region correspondences are derived (one geometric, the other photometric). They allow to test the consistency of correspondences and hence to reject falsely matched regions.

A Compact Model for Viewpoint Dependent Texture Synthesis

Abstract: A texture synthesis method is presented that generates similar texture from an example image. It is based on the emulation of simple but rather carefully chosen image intensity statistics. The resulting texture models are compact and no longer require the example image from which they were derived. They make explicit some structural aspects of the textures and the modeling allows knitting together different textures with convincingly looking transition zones. As textures are seldom flat, it is important to also model 3D effects when textures change under changing viewpoint. The simulation of such changes is supported by the model, assuming examples for the different viewpoints are given.

Augmented Reality Using Uncalibrated Video Sequences

Abstract: Augmented Reality (AR) aims at merging the real and the virtual in order to enrich a real environment with virtual information. Augmentations range from simple text annotations accompanying real objects to virtual mimics of real-life objects inserted into a real environment. In the latter case the ultimate goal is to make it impossible to differentiate between real and virtual objects. Several problems need to be overcome before realizing this goal. Amongst them are the rigid registration of virtual objects into the real environment, the problem of mutual occlusion of real and virtual objects and the extraction of the illumination distribution of the real environment in order to render the virtual objects with this illumination model. This paper will unfold how we proceeded to implement an Augmented Reality System that registers virtual objects into a totally uncalibrated video sequence of a real environment that may contain some moving parts. The other problems of occlusion and illumination will not be discussed in this paper but are left as future research topics.

Realistic surface reconstruction of 3D scenes from uncalibrated image sequences

Abstract: This contribution addresses the problem of obtaining 3D models from image sequences. A 3D surface description of the scene is extracted completely from a set of uncalibrated camera images of the scene. No prior knowledge about the scene or about the camera is needed to build the 3D models. The only assumptions are the rigidity of the scene objects and opaque object surfaces. The modelling system described here uses a three-step approach. First, the camera pose and intrinsic parameters are calibrated by tracking salient feature points throughout the sequence. Next, consecutive images of the sequence are treated as stereoscopic image pairs, and dense correspondence maps are computed by area matching. Finally, dense and accurate depth maps are computed by linking together all correspondences over the viewpoints. The depth maps are converted to triangular surfaces meshes that are texture mapped for photo-realistic appearance. The feasibility of the approach has been tested on both real and synthetic data and is illustrated here on several outdoor image sequences. Copyright © 2000 John Wiley & Sons, Ltd.

Three-dimensional scene reconstruction from images

Abstract: Modeling of 3D objects from image sequences is a challenging problem and has been a research topic for many years. Important theoretical and algorithmic results were achieved that allow to extract even complex 3D models of scenes from sequences of images. One recent effort has been to reduce the amount of calibration and to avoid restrictions on the camera motion. In this contribution an approach is described which achieves this gaol by coming state-of-the-art algorithms for uncalibrated projective reconstruction, self- calibration and dense correspondence matching.© (2000) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Minimal information to determine affine shape equivalence.

Abstract: Participants judged the affine equivalence of 2 simultaneously presented 4-point patterns. Performance level (d') varied between 1.5 and 2.7, depending on the information available for solving the correspondence problem (insufficient in Experiment la, superfluous in Experiment lb, and minimal in Experiments lc, 2a, 2b) and on the exposure time (unlimited in Experiments 1 and 2a and 500 ms in Experiment 2b), but it did not vary much with the complexity of the affine transformation (rotation and slant in Experiment 1 and same plus tilt in Experiment 2). Performance in Experiment 3 was lower with 3-point patterns than with 4-point patterns, whereas blocking the trials according to the affine transformation parameters had little effect. Determining affine shape equivalence with minimal-information displays is based on a fast assessment of qualitatively or quasi-invariant properties such as convexity/ concavity, parallelism, and collinearity.

Acquisition of detailed models for virtual reality

Abstract: More and more archaeological sites are being reconstructed in virtual reality. The problem remains the huge effort that has to be made to obtain realistic models. Besides on-site measurements, much time is often spent in manually rebuilding the whole site with a CAD package or a 3D-modelling tool. This limits the tractable complexity. In this paper two flexible automatic 3D surface acquisition systems are used to “virtualise” archaeological sites. The Roman site of Sagalassos (Southwest Turkey) is used as a test case to illustrate the potential of these new approaches. Besides the construction of a virtual site consisting of different level of details, some more applications to archaeology and conservation of heritage sites are presented.

Automatic 3d modeling from image sequences

Abstract: Modeling of three-dimensional (3D) objects from image sequences is a challenging problem and has been a research topic for many years. Important theoretical and algorithmic results were achieved that allow to extract even complex 3D models of scenes from sequences of images. One recent effort has been to reduce the amount of calibration and to avoid restrictions on the camera motion. In this contribution an approach is described which achieves this goal by combining state-of-the-art algorithms for uncalibrated projective reconstruction, self-calibration and dense correspondence matching.

Exploiting color for edge extraction and line segment stereo matching in high-resolution aerial imagery

Abstract: This paper investigates into the added value of color information for edge extraction and straight edge segment matching between stereo views. For edge extraction in color images different methods proposed in the literature are evaluated and compared, paying special attention to significance and completeness of the obtained edge-map. To find related edge segment pairs in different views, we apply an odd-man-out scheme: starting with all geometrically possible pairs we first rule out pairs, for which the chromatic information provided by the regions flanking the edge segments is incompatible. To further restrict the number of pairs we compute a chromatic similarity measure based on cross-correlation in the color bands. Both steps result in a significant reduction of candidate pairs, yet no correct pairs get lost. A main application of our technique is for automatic 3D building reconstruction from high resolution aerial images.

Exploiting color for edge extraction and line segment stereo matching

Abstract: This paper investigates into the added value of color information for edge extraction and straight edge segment matching between stereo views. For edge extraction in color images different methods proposed in the literature are evaluated and compared, paying special attention to significance and completeness of the obtained edge-map. To find related edge segment pairs in different views, we apply an odd-man-out scheme: starting with all geometrically possible pairs we first rule out pairs, for which the chromatic information provided by the regions flanking the edge segments is incompatible. To further restrict the number of pairs we compute a chromatic similarity measure based on cross-correlation in the color bands. Both steps result in a significant reduction of candidate pairs, yet no correct pairs get lost. A main application of our technique is for automatic 3D building reconstruction from high resolution aerial images.

Some issues on self-calibration and critical motion sequences

Abstract: This paper is concerned with the problem of selfcalibration. First some general concepts are discussed, then several methods are briefly discussed. In the second part of the paper the problem of critical motion sequences is treated. This problem could cause an important limitation to the practical use of self-calibration. In this context several interesting results are presented. self-calibration, critical motion sequences, absolute conic, ambiguous reconstruction.

Image-based 3D modeling: modeling from reality

Abstract: Increasingly, realistic object, scene, and event modeling is based on image data rather than manual synthesis The paper describes a system for visits to a virtual, 3D archeological site One can navigate through this environment, with a virtual guide as companion One can ask questions using natural, fluent speech The guide will respond and will bring the visitor to the desired place Simple answers are given as changes in the orientations of his head, by him raising his eyebrows or by head nodding In the near future the head will speak

Lip animation based on observed 3D speech dynamics

Abstract: We are all experts in the perception and interpretation of faces and their dynamics. This makes facial animation a particularly demanding area of graphics. Increasingly, computer vision is brought to bear and 3D models and their motions are learned from observations. The paper subscribes to this strand for the 3D modeling of human speech. The approach follows a kind of bootstrap procedure. First, 3D shape statistics are learned from faces with a few markers. A 3D reconstruction of a speaking face is produced for each video frame. A topological mask of the lower half of the face is fitted to the motion. The 3D shape statistics are extracted and principal components analysis reduces the dimension of the maskspace. The final speech tracker can work without markers, as it is only allowed to roam this constrained space of masks. Upon the representation of the different visemes in this space, speech or text can be used as input for animation.© (2000) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

to appear proceedings SPIE Electronic Imaging 2000, Three-Dimensional Image Capture and Applications III

Abstract: Modeling of three-dimensional (3D) objects from image sequences is a challenging problem and has been a research topic for many years. Important theoretical and algorithmic results were achieved that allow to extract even complex 3D models of scenes from sequences of images. One recent effort has been to reduce the amount of calibration and to avoid restrictions on the camera motion. In this contribution an approach is described which achieves this goal by combining state-of-the-art algorithms for uncalibrated projective reconstruction, self-calibration and dense correspondence matching.

Some geometric insight in self-calibration and critical motion sequences

Abstract: Self-calibration and critical motion sequences are problems which have received a lot of attention lately. The key concept for these problems is the absolute conic which appears as a natural calibration object always present in space. To observe it, however, self-calibration constraints are needed. In this paper a method to represent the rather abstract concept of the absolute conic through real geometric entities is proposed. The main beneet of this approach is that it allows to apply geometric intuition to understand problems related to self-calibration and critical motion sequences. The image of the absolute conic can be represented as an ellipse in the image. In this case self-calibration constraints are translated to simple geometric constraints on this ellipse. Several researchers have recently worked on critical motion sequences for self-calibration. The problem is however that the obtained results are hard to understand. Some cases are very hard to grasp intuitively and in an eeort to do so the proposed method was developed. By mapping the problem to real geometric entities, even the most intricate cases obtained by the previous analyses (often through the help of automated solving tools) can be understood and visualized. Several of these cases are discussed explicitly in this paper. Several new insights are also presented.

Transformations of a 3D Image Reconstruction Algorithm for Data Transfer and Storage Optimisation

Abstract: When implementing a 3D image reconstruction algorithm on a DSP architecture, we find ourselves confronted with a large memory transfer overhead, reducing the possible speedup attainable on recent multi-media oriented architectures. This paper describes how the critical part of the algorithm is re-specified and aggressively transformed at the algorithm code level, to improve the data access locality of the multi-dimensional image signal, while preserving the input/output behaviour. Experiments show that a close to optimal reuse of the data in the foreground memory and registers is obtained, removing the data transfer and storage bottleneck and enabling real-time prototyping of the algorithm on a DSP architecture.

Design and implementation of a supervised intelligent function

Abstract: This paper presents a generic design and implementation of one of the most useful functions required by intelligent living systems. The described system is capable of using an autonomous strategy for searching modelled objects in its environment , localising their positions, evaluating the safety of the intended actions by anticipating them in virtual worlds, and nally moving towards them for compliant contact and manipulation. To work in all conditions, the system enables a exible multi-modal interaction and a gradual human intervention , cooperation and co-assistance with the machine. Our experiments, carried out with a 6 degrees-of-freedom robot, deenitely connrm the possibility to use Integrated Supervisory Control systems as adaptive, safe and eecient synergetic human-machine systems for coping with various real-world applications.