Publications
[Complete list of all entries in BibTex format]
Dissertation: Model-Based High-Dimensional Pose Estimation with Application to Hand Tracking
Daniel Mohr
This thesis presents several novel techniques for computer vision based full-DOF human hand motion estimation. The most important contributions are a novel resolution-independent and memory efficient representation of hand pose silhouettes that allows to match a hypothesis in near-constant time, a new class of similarity measures that work for nearly arbitrary input modalities, and a novel matching approach that naturally combines a novel template hierarchy with a new image space search method.
Published:
Staats- und Universitätsbibliothek Bremen [BibTex]
Files:
Dissertation (pdf, 25MB)
New Geometric Data Structures for Collision Detection
Rene Weller
The dissertation of Rene Weller, including kinetic collision detection, Protosphere, Inner Sphere Trees, Sphere-Spring-Systems, benchmarking of collision detection algorithms and applications to haptics and robotics.
Published:
Staats- und Universitätsbibliothek Bremen [BibTex]
Files:
Paper (pdf, 50MB)
User Performance in Complex Bi-manual Haptic Manipulation with 3 DOFs vs. 6 DOFs
Rene Weller, Gabriel Zachmann
We present the results of a comprehensive user study that evaluates the influence of the degrees of freedom on the users' performance in complex bi-manual haptic interaction tasks. To do that, we have developed a novel multi-player game that allows the qualitative as well as the quantitative evaluation of different force-feedback devices simultaneously. The game closely resembles typical tasks arising in tele-operation scenarios or virtual assembly simulations; thus, the results of our user study apply directly to real-world industrial applications. The game is based on our new haptic workspace that supports high fidelity, two-handed multi-user interactions in scenarios containing a large number of dynamically simulated rigid objects; moreover, it works independent of the objects' polygon count. The results of our user study show that 6 DOF forcefeedback devices outperform 3 DOF devices significantly, both in user perception and in user performance.
Published:
Haptics Symposium 2012 [BibTex]
Files:
Paper (pdf, 2.6MB)
Poster (pdf, 1.6MB)
Eyecatcher (pdf, 0.8MB)
Teaser (pptx, 10.0MB) Best Teaser Award
Links:
A Comparative Evaluation of Three Skin Color Detection Approaches
Dennis Jensch, Daniel Mohr and Gabriel Zachmann
Skin segmentation is a challenging task due to several influences such as, for example, unknown lighting conditions, skin colored background, and camera limitations. A lot of skin segmentation approaches were proposed in the past including adaptive (in the sense of updating the skin color online) and non-adaptive approaches. In this paper, we compare three different skin segmentation approaches. The first is a well-known non- adaptive approach. It is based on a simple, pre-computed skin color distribution. Methods two and three adaptively estimate the skin color in each frame utilizing clustering algorithms. The second approach uses a hierarchical clustering for a simultaneous image and color space segmentation, while the third approach is a pure color space clustering, but with a more sophisticated clustering approach.
For evaluation, we compared the segmentation results of the approaches against a ground truth dataset. To obtain the ground truth dataset, we labeled about 500 images captured under various conditions.
Published:
In GI AR/VR Workshop 2012 [BibTex]
To appear in Journal of Virtual Reality and Broadcasting [BibTex]
Files:
Links:
Segmentation-Free, Area-Based Articulated Object Tracking
Daniel Mohr, Gabriel Zachmann
We propose a novel, model-based approach for articulated object detection and pose estimation that does not need any low-level feature extraction or foreground segmentation and thus eliminates this error-prone step. Our approach works directly on the input color image and is based on a new kind of divergence of the color distribution between an object hypothesis and its background. Consequently, we get a color distribution of the target object for free.
We further propose a coarse-to-fine and hierarchical algorithm for fast object localization and pose estimation. Our approach works significantly better than segmentation-based approaches in cases where the segmen- tation is noisy or fails, e.g. scenes with skin-colored backgrounds or bad illumination that distorts the skin color.
We also present results by applying our novel approach to markerless hand tracking.
Published:
7th International Symposium on Visual Computing (ISVC) 2011 [BibTex]
Files:
Paper
Slides [pptx]
[pdf]
Example Video
Links:
Adaptive Bitonic Sorting
Gabriel Zachmann
Adaptive bitonic sorting is a sorting algorithm suitable for implementation on EREW parallel architectures. Similar to bitonic sorting, it is based on merging, which is recursively applied to obtain a sorted sequence. In contrast to bitonic sorting, it is data-dependent. Adaptive bitonic merging can be performed in O(n/p) parallel time, p being the number of processors, and executes only O(n) operations in total. Consequently, adaptive bitonic sorting can be performed in O(n log n / p) time, which is optimal. So, one of its advantages is that it executes a factor of O(log n) less operations than bitonic sorting. Another advantage is that it can be implemented efficiently on modern GPUs.
Published:
Encyclopedia of Parallel Computing, Springer, 2011, pages 146-157; Padua, David (ed.) [BibTex]
3-DOF vs. 6-DOF - Playful Evaluation of Complex Haptic Interactions
Rene Weller, Gabriel Zachmann
We present a haptic workspace that allows high fidelity two-handed multi-user interactions in scenarios containing a large number of dynamically simulated rigid objects and a polygon count that is only limited by the capabilities of the graphics card. Based in this workspace we present a novel multiplayer game that supports qualitative as well as quantitative evaluation of different haptic devices in demanding haptic interaction tasks.
Published:
IEEE International Conference on Consumer Electronics (ICCE) 2011 [BibTex]
Files:
Paper (pdf, 0.5MB)
Slides (pptx, 15MB)
Video from Talk: wmv (60MB),
mov (76MB)
Links:
Inner Sphere Trees and Their Application to Collision Detection
Rene Weller, Gabriel Zachmann
Collision detection between rigid objects plays an important role in many fields of robotics and computer graphics, e.g. for path-planning, haptics, physically-based simulations, and medical applications.
This chapter contributes the following novel ideas to the area of collision detection:
- A novel geometric data structure, the Inner Sphere Trees (IST), that provides hierarchical bounding volumes from the inside of an object.
- A method to compute a dense sphere packing inside a polygonal object.
- We propose to utilize a clustering algorithm to construct a sphere hierarchy.
- A unified algorithm that can compute for a pair of objects, based on their ISTs, both an approximate minimal distance and the approximate penetration volume; the application does not need to know in advance which situation currently exists between the pair of objects.
- A method to compute forces from the penetration volume that are continuous, both in direction and value.
Published:
Virtual Realities, Springer, 2011 [BibTex]
Links:
ProtoSphere: A GPU-Assisted Prototype Guided Sphere Packing Algorithm for Arbitrary Objects
Rene Weller, Gabriel Zachmann
We present a new algorithm that is able to efficiently compute a space filling sphere packing for arbitrary objects. It is independent of the object's representation (polygonal, NURBS, CSG,...); the only precondition is that it must be possible to compute the distance from any point to the surface of the object. Moreover, our algorithm is not restricted to 3D but can be easily extended to higher dimensions.
The basic idea is very simple and related to prototype based approaches known from machine learning. This approach directly leads to a parallel algorithm that we have implemented using CUDA. As a byproduct, our algorithm yields an approximation of the object's medial axis that has applications ranging from path-planning to surface reconstruction.
Published:
Siggraph Asia, Technical Sketches, 2010 [BibTex]
Files:
Paper (pdf, 6MB)
Slides (pptx, 25MB)
Video 1 from Talk: wmv (30MB), mov (16MB)
Video 2 from Talk: wmv (8MB), mov (7MB)
Video 3 from Talk: wmv (2MB), mov (2MB)
Video 4 from Talk: wmv (5MB), mov (4MB)
Video 5 from Talk: wmv (4MB), mov (3MB)
Video 6 from Talk: wmv (28MB), mov (14MB)
Links:
A Benchmarking Suite for 6-DOF Real Time Collision Response Algorithms
Rene Weller, David Mainzer, Gabriel Zachmann, Mikel Sagardia, Thomas Hulin, Carsten Preusche
A benchmarking suite for rigid object collision detection and collision response schemes. The proposed benchmarking suite can evaluate both the performance as well as the quality of the collision response. The former is achieved by densely sampling the configuration space of a large number of highly detailed objects; the latter is achieved by a novel methodology that comprises a number of models for certain collision scenarios. With these models, we compare the force and torque signals both in direction and magnitude.
Our device-independent approach allows objective predictions for physically-based simulations as well as 6-DOF haptic rendering scenarios. In the results, we show a comprehensive example application of our benchmarks comparing two quite different algorithms utilizing our proposed benchmarking suite. This proves empirically that our methodology can become a standard evaluation framework.
Published:
Proceedings of the 17th ACM Symposium on Virtual Reality Software and Technology 2010 (VRST' 2010), Hong Kong, November 2010. [BibTex]
Files:
Paper (pdf, 2.2MB)
Presentation Slides (pdf, 1.2MB)
Links:
FAST: Fast Adaptive Silhouette Area based Template Matching
Daniel Mohr, Gabriel Zachmann
Template matching is a well-proven approach in the area of articulated object tracking. Matching accuracy and computation time of template matching are essential and yet often conflicting goals.
In this paper, we present a novel, adaptive template matching approach based on the silhouette area of the articulated object. With our approach, the ratio between accuracy and speed simply is a modifiable parameter, and, even at high accuracy, it is still faster than a state-of-the-art approach. We approximate the silhouette area by a small set of axis-aligned rectangles. Utilizing the integral image, we can thus compare a silhouette with an input image at an arbitrary position independently of the resolution of the input image. In addition, our rectangle covering yields a very memory efficient representation of templates.
Furthermore, we present a new method to build a template hierarchy optimized for our rectangular representation of template silhouettes. %This is a consistent continuation of our adaptive approach.
With the template hierarchy, the complexity of our matching method for n templates is O(log n) and independent of the input resolution. For example, a set of 3000 templates can be matched in 2.3 ms.
Overall, our novel methods are an important contribution to a complete system for tracking articulated objects.
Published:
British Machine Vision Conference 2010 [BibTex]
Files:
Links:
Silhouette Area Based Similarity Measure for Template Matching in Constant Time
Daniel Mohr, Gabriel Zachmann
We present a novel, fast, resolution-independent silhouette area-based matching approach. We approximate the silhouette area by a small set of axis-aligned rectangles. This yields a very memory efficient representation of templates. In addition, utilizing the integral image, we can thus compare a silhouette with an input image at an arbitrary position in constant time.
Furthermore, we present a new method to build a template hierarchy optimized for our rectangular representation of template silhouettes. With the template hierarchy, the complexity of our matching method for n templates is O(log n). For example, we can match a hierarchy consisting of 1000 templates in 1.5 ms. Overall, our contribution constitutes an important piece in the initialization stage of any tracker of (articulated) objects.
Published:
6th International Conference of Articulated Motion and Deformable Objects [BibTex]
The original publication is available at Springer Verlag
Files:
Paper Erratum: Eq. 5 does not take into account all rectangle configurations, i.e. we do not obtain the minimum number of rectangles for all areas.
Presentation Slides
Example Video
Links:
Stable 6-DOF Haptic Rendering with Inner Sphere Trees
Rene Weller, Gabriel Zachmann
Based on our new geometric data structure, the inner sphere trees, we present a fast and stable uniform algorithm for proximity and penetration volume queries between watertight objects at haptic rates.
Moreover, we present a multi-threaded version of the penetration volume computation for time-critical haptic rendering that is based on separation lists and the novel notion of expected overlapping volumes. Finally, we show how to use the penetration volume to compute continuous contact forces and torques that enable a stable rendering of 6-DOF penalty-based distributed contacts.
Published:
Proceedings of International Design Engineering Technical Conferences & Computers and Information in Engineering Conference (IDETC/CIE) , San Diego, USA, 30 August - 02 September 2009. Virtual Environments and Systems - 2009 Best Paper Award. [BibTex]
Files:
IDETC/CIE Presentation (ppt, 15MB)
Simulation Video wmv, mov
Interaction Video wmv, mov
Bones Video wmv, mov
758 Video wmv, mov
Pin in Hole Benchmark Video wmv, mov
Links:
Visual Computing for Medical Diagnosis and Treatment
Jan Klein, Ola Friman, Markus Hadwiger, Bernhard Preim, Felix Ritter, Anna Vilanova, Gabriel Zachmann, Dirk Bartz
Diagnostic algorithms and efficient visualization techniques are of major importance for pre-operative decisions, intra-operative imaging and image-guided surgery. Complex diagnostic decisions are characterized by a high information flow and fast decisions, requiring efficient and intuitive presentation of complex medical data and precision in the visualization. For intra-operative medical treatment, the pre-operative visualization results of the diagnostic systems have to be transferred to the patient on the operation room table. Via augmented reality, additional information of the hidden regions can be displayed virtually. This state-of-the-art report summarizes visual computing algorithms for medical diagnosis and treatment. After starting with direct volume rendering and tagged volume rendering as general techniques for visualizing anatomical structures, we go into more detail by focusing on the visualization of tissue and vessel structures. Afterwards, algorithms and techniques that are used for medical treatment in the context of image-guided surgery, intra-operative imaging and augmented reality, are discussed and reviewed.
Published:
Computers & Graphics, Vol. 33, Issue 4, August 2009, pp. 554 -- 565. [BibTex]
Files:
Preliminary version of the paper (pdf),
A Unified Approach for Physically-Based Simulations and Haptic Rendering
Rene Weller, Gabriel Zachmann
Since the visual feedback and effects of today's games have become extremely mature, it will be more and more important for games to provide realistic feedback to other senses, such as our haptic sense. On the hardware side, this has become possible in recent years by the advent of first inexpensive haptic devices on the consumer market, such as the Falcon from Novint. Research on force-feedback devices and algorithms has been done over 10 years, and has only fairly recently been introduced to games.
However, while there is a large body of research on how to render forces given a collision and its contact information, the computation of the latter for massive models is still a challenge. First of all, this is due to the much higher effort to compute contact information. Second, this is due to the update rates that are necessary for haptic rendering, which need to be much higher than for visual rendering, i.e., 250-1000 Hz. And third, defining the contact information such that continuous contact forces can be derived is not always obvious.
Therefore, one of the major challenges in haptic rendering for games is the computation of continuous forces at haptic rates. A solution to this challenge can also be utilized to do physically-based simulation of rigid bodies, which has become increasingly popular in games over the past few years.
In this paper, we take advantage of the fact that in rendering haptic forces, as well as in most real-time applications that involve physically-based simulation, an absolutely correct determination of the forces acting on the virtual objects is not necessary.
Published:
ACM SIGGRAPH Video Game Proceedings , New Orleans, USA, August 2009. [BibTex]
Files:
Siggraph Paper (pdf, 4MB)
Siggraph Presentation (ppt, 25MB)
Simulation Video wmv, mov
Interaction Video wmv, mov
Armadillo Video wmv, mov
Screwdriver Video wmv, mov
Bozzle Video wmv, mov
Links:
Inner Sphere Trees for Proximity and Penetration Queries
Rene Weller, Gabriel Zachmann
We present a novel geometric data structure for approximate collision detection at haptic rates between rigid objects. Our data structure, which we call inner sphere trees, supports different kinds of queries, namely, proximity queries and a new method for interpenetration computation, the penetration volume, which is related to the water displacement of the overlapping region and, thus, corresponds to a physically motivated force. The main idea is to bound objects from the inside with a set of non-overlapping spheres. Based on such sphere packings, a "inner bounding volume hierarchy" can be constructed. In order to do so, we propose to use an AI clustering algorithm, which we extend and adapt here. The results show performance at haptic rates both for proximity and penetration volume queries for models consisting of hundreds of thousands of polygons.
Published:
2009 Robotics: Science and Systems Conference (RSS) , Seattle, USA, 28 June - 01 July 2009. [BibTex]
Files:
RSS paper (pdf, 6MB),
RSS poster (pdf, 2MB)
Technical Report (pdf, 6MB), [BibTex]
Links:
Continuous Edge Gradient-Based Template Matching for Articulated Objects
Gabriel Zachmann, Daniel Mohr
In this paper, we propose a novel edge gradient based template matching method for object detection. In contrast to other methods, ours does not perform any binarization or discretization during the online matching. This is facilitated by a new continuous edge gradient similarity measure. Its main components are a novel edge gradient operator, which is applied to query and template images, and the formulation as a convolution, which can be computed very efficiently in Fourier space.
Published
International Conference on Computer Vision Theory and Applications (VISAPP) , Lisbon, Portugal, 05-09 February 2009. [BibTex]
Files:
Paper
Slides
Technical Report
Video 1divx,
Video 2divx,
Video 3divx
Links:
Segmentation of Distinct Homogeneous Color Regions in Images
Gabriel Zachmann, Daniel Mohr
In this paper, we present a novel algorithm to detect homogeneous color regions in images with application to skin segmentation.
Published
The 12th International Conference on Computer Analysis of Images and Patterns (CAIP), Vienna, Austria, 27-29 August 2007. [BibTex]
Files:
Paper (pdf, 1.6MB)
Slides (pdf, 6.2MB)
Video (divx, 2.2MB)
Video (mov, 3.5MB)
Links:
IEEE VR2007 Workshop on "Trends and Issues in Tracking for Virtual Environments"
Gabriel Zachmann (ed.)
The goal of this half-day workshop is to bring together researchers and industry working in the area of tracking and to talk about making tracking actually work. To that end, the workshop is to provide a broad picture of what is the current state of the art, what are the various technologies available, and what are issues for further research and development
Published
February 2007, Shaker Verlag, Aachen, Germany, ISBN 978-3-8322-5967-9 [BibTeX]
Files:
Home page of the Workshop
Buy
a copy of the the workshop proceedings online (hard-copy for 25€;
electronic version for 3€) from the publisher
You can also
ask me at
zach at informatik.uni-bremen.de
-- I've still got some copies for sale, left over from the conference ;-)
A Benchmarking Suite for Static Collision Detection Algorithms
Sven Trenkel, René Weller, Gabriel Zachmann
In this paper, we present a benchmarking suite that allows a systematic comparison of pairwise static collision detection algorithms for rigid objects. The pdf-file contains a slightly extended version of the original published paper.
Published
International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG), Plzen, Czech Republic, January 29 - February 1, 2007. [BibTex]
Files:
Paper (pdf, 1.0MB)
Slides (pdf, 5.4MB)
Video of configuration generation (avi, 16.6MB)
For further information please visit our project homepage.
Kinetic Separation Lists for Continuous Collision Detection of Deformable Objects
Gabriel Zachmann, René Weller
In this paper, we present a new acceleration scheme for continuous inter- and intra-collision detection of deformable objects. The pdf-file contains a slightly extended version of the original published paper (Section 6.2).
Published
Third Workshop in Virtual Reality Interactions and Physical Simulation (Vriphys), Madrid, Spain, 6/7 November 2006. [BibTex]
Files:
Paper (pdf, 0.7MB)
Slides (pdf, 4.9MB)
Kinetic Bounding Volume Hierarchies for Deformable Objects
Gabriel Zachmann, René Weller
In this paper, we present novel algorithms for updating bounding volume hierarchies of objects undergoing arbitrary deformations.
Published
ACM Int'l Conf. on Virtual Reality Continuum and Its Applications (VRCIA), Hong Kong, China, 14-17 June 2006. [BibTex]
Files:
Paper (pdf, 1MB)
Slides (pdf, 6.0 MB)
Technical Report (pdf, 1.2MB)
A Model for the Expected Running Time of Collision Detection using AABB Trees
René Weller, Jan Klein, Gabriel Zachmann
In this paper, we propose a model to estimate the expected running time of hierarchical collision detection that utilizes AABB trees, which are a frequently used type of bounding volume (BV).
Published
12th Eurographics Symposium on Virtual Environments (EGVE), Lisbon, Portugal, 8-10 May 2006. [BibTex]
Files:
Paper, on-screen version (pdf, 1.6 MB)
Paper, print version (pdf, 1.6 MB)
Slides (pdf, 2.5 MB)
Book: Geometric Data Structures for Computer Graphics
Elmar Langetepe, Gabriel Zachmann, AK Peters, 2006
Links
Buy from the publisher or from Amazon BibTex
GPU-ABiSort: Optimal Parallel Sorting on Stream Architectures
Alexander Greß, Gabriel Zachmann
In this paper, we present a novel approach for parallel sorting on stream processing architectures.
Published
Proc. 20th IEEE International Parallel and Distributed Processing Symposium (IPDPS), Rhodes Island, Greece, 25-29 April 2006. [BibTex]
Files:
Paper (pdf, 0.3MB)
Technical Report (pdf, 1.3MB)
Space-Efficient FPGA-Accelerated Collision Detection for Virtual Prototyping
Andreas Raabe, Stefan Hochgürtel, Gabriel Zachmann, Joachim K. Anlauf
In this paper, we present a space-efficient, FPGA-optimized architecture to detect collisions among virtual objects, as well as simulation results for collision queries using this architecture.
Published
Design Automation and Test in Europe (DATE), Munich, Germany, 6-10 March 2006. [BibTex]
Files:
Paper (pdf, 0.3MB)
Slides (pdf, 0.8MB)
Hardware-Accelerated Collision Detection using Bounded-Error Fixed-Point Arithmetic
Andreas Raabe, Stefan Hochgürtel, Gabriel Zachmann, Joachim K. Anlauf
In this paper, we present a novel approach for highly space-efficient hardware-accelerated collision detection.
Published
The 14-th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG), Plzen, Czech Republic, 30 January - 3 February 2006. [BibTex]
Files:
Paper (pdf, 0.4MB)
Slides (pdf, 0.6MB)
Patent on Collision Detection
Gabriel Zachmann, US Patent and Trademark Office, May 2005, [BibTex]
Links:
The Expected Running Time of Hierarchical Collision Detection
Jan Klein, Gabriel Zachmann
We propose a theoretical approach to analyze the average-case running time of hierarchical collision detection that utilizes bounding volume hierarchies.
Published
SIGGRAPH 2005, Poster, Los Angeles, CA, USA, August 2005. [BibTex]
Files:
Poster (pdf, 1.8MB)
One-Page Summary. (pdf, 1.6 MB)
Supplemental Material (pdf, 1.4 MB)
Slides (pdf, 0.6 MB)
Collision Detection for Deformable Objects
M. Teschner, S. Kimmerle, B. Heidelberger, G. Zachmann, L. Raghupathi, A. Fuhrmann, M.-P. Cani, F. Faure, N. Magnenat-Thalmann, W. Strasser, P. Volino
This paper focusses on collision detection for deformable objects and summarizes recent research in the area of deformable collision detection. Various approaches based on bounding volume hierarchies, distance fields, and spatial partitioning are discussed. Further, image-space techniques and stochastic methods are considered. Applications in cloth modeling and surgical simulation are presented.
Published
Computer Graphics forum, March 2005, Volume 24 [BibTex]
Files:
Paper (gzipped postscript, 2.8MB)
Paper (pdf, 10MB)
Slides (pdf, 0.5MB)
Interpolation Search for Point Cloud Intersection
Jan Klein, Gabriel Zachmann
We present a novel algorithm to compute intersections of two point clouds.
Published
Proc. of WSCG 2005, Plzen, Czech Republic, 31 January - 7 February 2005. [BibTex]
Files:
Paper (pdf print-version, 2.4MB)
Paper (pdf onscreen-version, 2.4MB)
Slides (pdf, 4.1MB)
Shader Maker
Markus Kramer, Rene Weller, Gabriel Zachmann
Actually, this is not a regular publication, but a software release.
Shader Maker is a simple, cross-platform GLSL editor. It works on Windows, Linux, and Mac OS X.
It provides the basics of a shader editor, such that students can get started with writing their own shaders as quickly as possible. This includes: syntax highlighting in the GLSL editors; geometry shader editor (as well as vertex and fragment shader editors, of course); interactive editing of the uniform variables; light source parameters; pre-defined simple shapes (e.g., torus et al.) and a simple OBJ loader; and a few more.
For download and further information please visit our project website