Mgr. Jana Kostlivá, Ph.D.

In this paper, we propose a new algorithm for fairing discrete surfaces resulting from stereo-based 3D reconstruction task. Such results are typically too dense, uneven and noisy, which is inconvenient for further processing. Our approach jointly optimises mesh smoothness and regularity. The definition is given on a discrete surface and the solution is found by discrete diffusion of a scalar function. Experiments on synthetic and real data demonstrate that the proposed approach is robust, stable, preserves qualitative shape and is applicable to even moderate-size real surfaces with boundary (0.8M vertices and 1.7M triangles).

In stereo literature, there is no standard method for evaluating algorithms for semi-dense stereo matching. Moreover, existing evaluations for dense methods require a fixed parameter setting for the tested algorithms. In this paper, we propose a method that overcomes these drawbacks and still is able to compare algorithms based on a simple numerical value, so that reporting results does not take up much space in a paper. We propose evaluation of stereo algorithms based on Receiver Operating Characteristics (ROC) which captures both errors and sparsity. By comparing ROC curves of all tested algorithms we obtain the Feasibility Boundary, the best possible performance achieved by a set of tested stereo algorithms, which allows stereo algorithm users to select the proper method and parameter setting for a required application.

We present an automatic pipeline for recovering the geometry of a 3D scene from a set of unordered, uncalibrated images. The contributions in the paper are the presentation of the system as a whole, from images to geometry, the estimation of the local scale for various scene components in the orientation-topology module, the procedure for orienting the cloud components, and the method for dealing with points of contact. The methods are aimed to process complex scenes and nonuniformly sampled, noisy data sets.

Known disparity search range is crucial for stereo matching tasks, since many algorithms require the disparity search range to be known. Searching over the whole disparity range (i.e. [-image width, image width]) is not only very time consuming (mainly for large images), but even many stereo algorithms do not perform well with unspecified disparity search range. Therefore, automatic estimation of disparity search range for unknown stereo image pairs is highly desired. In low-level image processing (i.e. without knowing any information about the captured scene) this task is very difficult. We propose an approach based on Confidently Stable Matching, which is fast, precise and robust. We demonstrate the algorithm properties on benchmark image sets with known disparity search range as well as on unknown complex scenes.

Recently it has been recognized that local image modeling in stereo matching brings more discriminable and stable matching features. We introduce a new disparity component matching approach, where the aggregation support regions are defined in disparity space, which guarantees invariance to input image view selection and independence on projective distortions. A rigorous ground-truth evaluation experiment showed that our approach is able to improve overall matching failure rate four-fold while the accuracy is preserved as compared to a standard fixed-size rectangular matching windows approach.

Recently it has been recognized that local image modeling in stereo matching brings more discriminable and stable matching features. We introduce a new disparity component matching approach, where the aggregation support regions are defined in disparity space, which guarantees invariance to input image view selection and independence on projective distortions. A rigorous ground-truth evaluation experiment showed that our approach is able to increase overall matching failure rate four-fold while the accuracy is preserved as compared to a standard fixed-size rectangular matching windows approach.

The knowledge of stereo matching algorithm properties and behaviour under varying conditions is crucial for the selection of a proper method for the desired application. In this paper we study the behaviour of four representative matching algorithms under varying signal-to-noise ratio in six types of error statistics. The errors are focused on basic matching failure mechanisms and their definition observes the principles of independence, symmetry and completeness. A ground truth experiment shows that the best choice for view prediction is the Graph Cuts algorithm and for structure reconstruction it is the Confidently Stable Matching.

Local joint image modeling in stereo matching brings more discriminable and stable matching features. Such features reduce the need for strong prior models (continuity) and thus algorithms that are less prone to false positive artefacts in general complex scenes can be applied. One of the principal quality factors in area-based dense stereo is the matching window shape. As it cannot be selected without having any initial matching hypothesis we propose a stratified matching approach. The window adapts to high-correlation structures in disparity space found in pre-matching which is then followed by final matching. In a rigorous ground-truth experiment we show that Stratified Dense Matching is able to increase matching density 3x, matching accuracy 1.8x, and occlusion boundary detection 2x as compared to a fixed-size rectangular windows algorithm. Performance on real outdoor complex scenes is also evaluated.

Stable Matching Based on Disparity Components