@inproceedings{LNCS58150305, editor = {Mario Fritz and Bernt Schiele and Justus H. Piater}, booktitle = {Computer Vision Systems}, publisher = {Springer}, location = {Heidelberg}, series = {Lecture Notes in Computer Science}, volume = {5815}, year = {2009}, isbn = {978-3-642-04666-7}, author = {Timo Dickscheid and Wolfgang F\"{o}rstner}, title = {Evaluating the Suitability of Feature Detectors for Automatic Image Orientation Systems}, pages = {305--314} }

@inproceedings{drauschke.2009.multidodging, author = "Drauschke, Martin and Förstner, Wolfgang and Brunn, Ansgar", title = "Multidodging: Ein effizienter Algorithmus zur automatischen Verbesserung von digitalisierten Luftbildern", pages = "61-68", city = "Bonn", proceeding = "Publikationen der DGPF, Band 18: Zukunft mit Tradition",}

@inproceedings{Drauschke-Roscher-Laebe-Foerstner_-_2009_-_Improving-Image-Segmentation-Using-Multi-View-Analysis, author = "Drauschke, Martin and Roscher, Ribana and Läbe, Thomas and Förstner, Wolfgang", title = "Improving Image Segmentation using Multiple View Analysis", pages = "211-226", city = "Paris", proceeding = "Object Extraction for 3D City Models, Road Databases and Traffic Monitoring - Concepts, Algorithms and Evaluatin (CMRT09)",}

@inproceedings{foerstner.2009.mustererkennung, author = "Förstner, Wolfgang", title = "Mustererkennung in der Fernerkundung", pages = "129-136", city = "Bonn", proceeding = "Publikationen der DGPF, Band 18: Zukunft mit Tradition",}

Photogrammetry has significantly been influenced by its two neigbouring fields, namely Computer Vision and Remote Sensing. Today, Photogrammetry has been become a part of Remote Sensing. The paper reflects its growing relations with Computer Vision, based on a more than 25 years experience of the author with the fascinating field between cognitive, natural and engineering science, which stimulated his own research and transferred him into a wanderer between two worlds.

@inproceedings{foerstner09.computervision, author = "Förstner, Wolfgang", title = "Computer Vision and Remote Sensing - Lessons Learned", pages = "241-249", city = "Stuttgart", proceeding = "Photogrammetric Week 2009",}

@inproceedings{foerstner*09:detecting, author={W. F\"orstner and T. Dickscheid and F. Schindler}, title={Detecting Interpretable and Accurate Scale-Invariant Keypoints}, booktitle={12th IEEE International Conference on Computer Vision (ICCV'09)}, address={Kyoto, Japan}, year={2009} pages={2256--2263} }

@inproceedings{foerstner*09:completeness, author={W. F\"orstner and T. Dickscheid and F. Schindler}, title={{On the Completeness of Coding with Image Features}}, booktitle={20th British Machine Vision Conference}, address={London, UK}, year={2009} }

We describe ground truth data that we provide to serve as a basis for evaluation and comparison of supervised learning approaches to image interpretation. The provided ground truth, the eTRIMS Image Database, is a collection of annotated images of real world street scenes. Typical objects in these images are variable in shape and appearance, in the number of its parts and appear in a variety of congurations. The domain of man-made scenes is thus well suited for evaluation and comparison of a variety of interpretation approaches, including those that employ structure models. The provided pixelwise ground truth assigns each image pixel both with a class label and an object label and offers thus ground truth annotation both on the level of pixels and regions. While we believe that such ground truth is of general interest in supervised learning, such data may be of further relevance in emerging real world applications involving automation of man-made scene interpretation.

@techreport{ korc-forstner-tr09-etrims, author = "Kor{\v c}, F. and F{\" o}rstner, W.", title = "{eTRIMS} {I}mage {D}atabase for Interpreting Images of Man-Made Scenes", number = "TR-IGG-P-2009-01", month = "April", year = "2009", institute = "Dept. of Photogrammetry, University of Bonn", url = "http://www.ipb.uni-bonn.de/projects/etrims_db/" }

Well known estimation techniques in computational geometry usually deal only with single geometric entities as unknown parameters and do not account for constrained observations within the estimation. The estimation model proposed in this paper is much more general, as it can handle multiple homogeneous vectors as well as multiple constraints. Furthermore, it allows the consistent handling of arbitrary covariance matrices for the observed and the estimated entities. The major novelty is the proper handling of singular observation covariance matrices made possible by additional constraints within the estimation. These properties are of special interest for instance in the calculus of algebraic projective geometry, where singular covariance matrices arise naturally from the non-minimal parameterizations of the entities. The validity of the proposed adjustment model will be demonstrated by the estimation of a fundamental matrix from synthetic data and compared to heteroscedastic regression [?], which is considered as state-ofthe- art estimator for this task. As the latter is unable to simultaneously estimate multiple entities, we will also demonstrate the usefulness and the feasibility of our approach by the constrained estimation of three vanishing points from observed uncertain image line segments.

@inproceedings{meidow*09:optimal, author={Meidow, J. and F{\" o}rstner, W. and Beder, C.}, title={{Optimal Parameter Estimation with Homogeneous Entities and Arbitrary Constraints}}, booktitle={Pattern Recognition (Symposium of DAGM) '09}, editor = {Denzler, J. and Notni, G.}, year={2009}, publisher = {Springer}, series = {LNCS} }

@TECHREPORT{Roscher2009, author = {Roscher, R., Foerstner, W.}, title = {{Multiclass Bounded Logistic Regression -- Efficient Regularization with Interior Point Method}}, institution = {Department of Photogrammetry, University of Bonn}, year = {2009}, number = {TR-IGG-P-2009-02} }

This report points out the role of sequences of samples for training an incremental learning method. We define characteristics of incremental learning methods to describe the influence of sample ordering on the performance of a learned model. Different types of experiments evaluate these properties for two different datasets and two different incremental learning methods. We show how to find sequences of classes for training just based on the data to get always best possible error rates. This is based on the estimation of Bayes error bounds.

@TECHREPORT{wenzel09*:role, author = {Susanne Wenzel and Wolfgang F{\"o}rstner}, title = {The Role of Sequences for Incremental Learning}, institution = {Department of Photogrammetry, University of Bonn}, year = {2009}, number = {TR-IGG-P-2009-04}, month = {October} }

In classical photogrammetry, point observations are manually determined by an operator for performing the bundle adjustment of a sequence of images. In such cases, a comparison of different estimates is usually carried out with respect to the estimated 3D object points. Today, a broad range of automatic methods are available for extracting and matching point features across images, even in the case of widely separated views and under strong deformations. This allows for fully automatic solutions to the relative orientation problem, and even to the bundle triangulation in case that manually measured control points are available. However, such systems often contain random subprocedures like RANSAC for eliminating wrong correspondences, yielding different 3D points but hopefully similar orientation parameters. This causes two problems for the evaluation: First, the randomness of the algorithm has an influence on its stability, and second, we are constrained to compare the orientation parameters instead of the 3D points. We propose a method for benchmarking automatic bundle adjustments which takes these constraints into account and uses the orientation parameters directly. Given sets of corresponding orientation parameters, we require our benchmark test to address their consistency of the form deviation and the internal precision and their precision level related to the precision of a reference data set. Besides comparing different bundle adjustment methods, the approach may be used to safely evaluate effects of feature operators, matching strategies, control parameters and other design decisions for a particular method. The goal of this paper is to derive appropriate measures to cover these aspects, describe a coherent benchmarking scheme and show the feasibility of the approach using real data.

@INPROCEEDINGS{dickscheid*08:benchmarking, author = {Dickscheid, Timo and L{\"a}be, Thomas and F{\" o}rstner, Wolfgang}, title = {Benchmarking Automatic Bundle Adjustment Results}, booktitle = {To appear in: 21st Congress of the International Society for Photogrammetry and Remote Sensing (ISPRS)}, address = {Beijing, China}, year = {2008}, }

@INPROCEEDINGS{drauschke*08:comparison, author = {Drauschke, Martin and F{\" o}rstner, Wolfgang}, title = {Comparison of Adaboost and ADTboost for Feature Subset Selection}, booktitle = {PRIS 2008}, address = {Barcelona, Spain}, year = {2008}, }

The paper addresses the problem of feature selection during classification of image regions within the context of interpreting images showing highly structured objects such as buildings. We present a feature selection scheme that is connected with the classification framework Adaboost, cf. (Schapire and Singer, 1999). We constricted our weak learners on threshold classification on a single feature. Our experiments showed that the classification with Adaboost is based on relatively small subsets of features. Thus, we are able to find sets of appropriate features. We present our results on manually annotated and automatically segmented regions from facade images of the eTRIMS data base, where our focus were the object classes facade, roof, windows and window panes.

@INPROCEEDINGS{drauschke*08:selecting, author = {Drauschke, Martin and F{\" o}rstner, Wolfgang}, title = {Selecting appropriate features for detecting buildings and building parts}, booktitle = {To appear in: 21st Congress of the International Society for Photogrammetry and Remote Sensing (ISPRS)}, address = {Beijing, China}, year = {2008}, }

We present a solution to the following discrete optimization problem. Given a set of independent, possibly overlapping image regions and a non-negative likeliness of the individual regions, we select a non-overlapping subset that is optimal with respect to the following requirements: First, every region is either part of the solution or has an overlap with it. Second, the degree of overlap of the solution with the rest of the regions is maximized together with the likeliness of the solution. Third, the likeliness of the individual regions influences the overall solution proportionally to the degree of overlap with neighboring regions. We represent the problem as a graph and solve the task by reduction to a constrained binary integer programming problem. The problem involves minimizing a linear objective function subject to linear inequality constraints. Both the objective function and the constraints exploit the structure of the graph. We illustrate the validity and the relevance of the proposed formulation by applying the method to the problem of facade window extraction. We generalize our formulation to the case where a set of hypotheses is given together with a binary similarity relation and similarity measure. Our formulation then exploits combination of degree and structure of hypothesis similarity and likeliness of individual hypotheses. In this case, we present a solution with non-similar hypotheses which can be viewed as a non-redundant representation.

@INPROCEEDINGS{korc*08:optimal, author = {Kor{\v c}, Filip and F{\" o}rstner, Wolfgang}, title = {Finding Optimal Non-Overlapping Subset of Extracted Image Objects}, booktitle = {Proc. of the 12th International Workshop on Combinatorial Image Analysis (IWCIA)}, year = {2008}, }

We investigate Discriminative Random Fields (DRF) which provide a principled approach for combining local discriminative classifiers that allow the use of arbitrary overlapping features, with adaptive data-dependent smoothing over the label field. We discuss the differences between a traditional Markov Random Field (MRF) formulation and the DRF model, and compare the performance of the two models and an independent sitewise classifier. Further, we present results suggesting the potential for performance enhancement by improving state of the art parameter learning methods. Eventually, we demonstrate the application feasibility on both synthetic and natural images.

@INPROCEEDINGS{korc*08:interpreting, author = {Kor{\v c}, Filip and F{\" o}rstner, Wolfgang}, title = {Interpreting Terrestrial Images of Urban Scenes Using Discriminative Random Fields}, booktitle = {Proc. of the 21st Congress of the International Society for Photogrammetry and Remote Sensing (ISPRS)}, year = {2008} }

We investigate maximum likelihood parameter learning in Conditional Random Fields (CRF) and present an empirical study of pseudo-likelihood (PL) based approximations of the parameter likelihood gradient. We show that these parameter learning methods can be improved and evaluate the resulting performance employing different inference techniques. We show that the approximation based on penalized pseudo-likelihood (PPL) in combination with the Maximum A Posteriori (MAP) inference yields results comparable to other state of the art approaches, while providing low complexity and advantages to formulating parameter learning as a convex optimization problem. Eventually, we demonstrate applicability on the task of detecting man-made structures in natural images.

@INPROCEEDINGS{korc*08:approximate, author = {Kor{\v c}, Filip and F{\" o}rstner, Wolfgang}, title = {Approximate Parameter Learning in {C}onditional {R}andom {F}ields: An Empirical Investigation}, booktitle = {Pattern Recognition}, year = {2008}, editor = {G. Rigoll}, number = {5096}, series = {LNCS}, pages = {11-20}, publisher = {Springer} }

This paper presents an empirical investigation into the quality of automatic relative orientation procedures. The results of an in-house developed automatic orientation software called aurelo (Laebe and Foerstner, 2006) are evaluated. For this evaluation a recently proposed consistency measure for two sets of orientation parameters (Dickscheid et. al., 2008) and the ratio of two covariances matrices is used. Thus we evaluate the consistency of bundle block adjustments and the precision level achievable. We use different sets of orientation results related to the same set of images but computed under differing conditions. As reference datasets results on a much higher image resolution and ground truth data from artificial images rendered with computer graphics software are used. Six different effects are analysed: varying results due to random procedures in aurelo, computations on different image pyramid levels and with or without points with only two or three observations, the effect of replacing the used SIFT operator with an approximation of SIFT features, called SURF, repetitive patterns in the scene and remaining non-linear distortions. These experiments show under which conditions the bundle adjustment results reflect the true errors and thus give valuable hints for the use of automatic relative orientation procedures and possible improvements of the software.

@INPROCEEDINGS{laebe*08:quality, author = {L{\"a}be, Thomas and Dickscheid, Timo and F{\" o}rstner, Wolfgang}, title = {On the Quality of Automatic Relative Orientation Procedures}, booktitle = {To appear in: 21st Congress of the International Society for Photogrammetry and Remote Sensing (ISPRS)}, address = {Beijing, China}, year = {2008}, }

This paper addresses the challenge of a real-time capable vision system in the task of trajectory and surface reconstruction by aerial image sequences. The goal is to present the design, methods and strategies of a real-time capable vision system solving the mapping task for secure navigation of small UAVs with a single camera. This includes the estimation process, map representation, initialization processes, loop closing detection and exploration strategies. The estimation process is based on the Kalman-Filter and a landmark based map representation. We introduce a new initialization method for new observed landmarks. We will show that the initialization process and the exploration strategy has a significant effect on the accuracy of the estimated camera trajectory and of the map.

@inproceedings{steffen08.visual, author = "Steffen, Richard and Förstner, Wolfgang", title = "On Visual Real Time Mapping for Unmanned Aerial Vehicles", pages = "57-62 Part B3a", city = "Bonn", proceeding = "21st Congress of the International Society for Photogrammetry and Remote Sensing (ISPRS)",}

We present a method for detecting repeated structures, which is applied on facade images for describing the regularity of their windows. Our approach finds and explicitly represents repetitive structures and thus gives initial representation of facades. No explicit notion of a window is used; thus, the method also appears to be able to identify other manmade structures, e.g., paths with regular tiles. A method for detection of dominant symmetries is adapted for detection of multiply repeated structures. A compact description of the repetitions is derived from the detected translations in the image by a heuristic search method and the criterion of the minimum description length.

@ARTICLE{wenzel*08:detection, author = {Wenzel, S. and Drauschke, M. and F{\" o}rstner, W.}, title = {Detection of repeated structures in facade images}, journal = {Pattern Recognition and Image Analysis}, year = {2008}, volume = {18}, pages = {406--411}, number = {3}, month = {September}, doi = {10.1134/S1054661808030073} }

We propose an incremental learning scheme for learning a class hierarchy for objects typically occurring multiple in images. Given one example of an object that appears several times in the image, e.g. is part of a repetitive structure, we propose a method for identifying prototypes using an unsupervised clustering procedure. These prototypes are used for building a hierarchical appearance based model of the envisaged class in a supervised manner. For classification of new instances detected in new images we use linear subspace methods that combine discriminative and reconstructive properties. The used methods are chosen to be capable for an incremental update. We test our approach on facade images with repetitive windows and balconies. We use the learned object models to find new instances in other images, e. g. the neighbouring facade and update already learned models with the new instances.

@INPROCEEDINGS{wenzel*08:semi-supervised, author = {Wenzel, Susanne and F{\" o}rstner, Wolfgang}, title = {Semi-supervised incremental learning of hierarchical appearance models}, booktitle = {21st Congress of the International Society for Photogrammetry and Remote Sensing (ISPRS)}, address = {Beijing, China}, year = {2008}, }