788 research outputs found
Bau der leichten Fahrzeuge und mobilen Brücken über Bäche, Flüsse und Sümpfe
No full textvon C. F. W. Schiele...; mit taktischen Anmerkungen von J. von NiedermayrKeine weiteren Bände erschienenSupralibros: "EA" 990005729620205503_0001 Exemplar der ETH-BI
Artistic presentation of seizure semiology—A remarkable painting
No full textZusammenfassung Ein Gemälde wird gezeigt, das die Anfallssemiologie der Patientin, Koautorin und Malerin Martina Schiele visualisiert. Es wird aufgezeigt, dass der künstlerische Akt half, das verbal Schwerbeschreibliche zu erfassen und darzustellen. Es erfolgen eine epileptologische Kontextualisierung und ein Ausblick auf Epilepsie und Kunst.A painting is shown that depicts the seizure semiology of a patient, i.e., Martina Schiele the painter and co-author of this article. It is illustrated that the artistic act helped to express and better understand her symptoms, which were hardly describable in words. A contextualisation and perspective on epilepsy and art is given.Open Access funding enabled and organized by Projekt DEAL.Universitätsklinikum Erlangen (8546
Video Segmentation with Superpixels
No full textDue to its importance, video segmentation has regained interest recently. However, there is no common agreement about the necessary ingredients for best performance. This work contributes a thorough analysis of various within- and between-frame affinities suitable for video segmentation. Our results show that a frame-based superpixel segmentation combined with a few motion and appearance-based affinities are sufficient to obtain good video segmentation performance. A second contribution of the paper is the extension of [1] to include motion-cues, which makes the algorithm globally aware of motion, thus improving its performance for video sequences. Finally, we contribute an extension of an established image segmentation benchmark [1] to videos, allowing coarse-to-fine video segmentations and multiple human annotations. Our results are tested on BMDS [2], and compared to existing methods
Jak drapieżny ptak chcę okrążyć miasto. Egon Schiele – melancholik skandalista
No full textEgon Schiele is one of the most distinctive painters of the early twentieth century. Known primarily as the author of scandalous acts. But also symbolic compositions, antropomorphic landscapes and still-life work for to his credit. The article presents the silhouette of the artist - egocentric, rebelious and eccentric. The artist who talks about his experience in a shamelessly open and honest way. The artist seeking knowledge about man in all of its abundance of uniqueness and weakness. The text is an attempt to answer the question about the phenomenon of the Viennese painter who courageously followed melancholic clues in one of the groundbreaking moments in the history - the decaying era.Egon Schiele to jeden z najbardziej charakterystycznych malarzy początku XX wieku. Znany przede wszystkim jako twórca skandalizujących aktów, ma w swoim dorobku również kompozycje symboliczne oraz pejzaże przedstawiające antropomorfizowane krajobrazy i martwe miasta. Artykuł przedstawia sylwetkę malarza – egocentryka, buntownika, ekscentryka – artysty mówiącego o swoim doświadczeniu w sposób bezwstydnie otwarty i szczery, szukającego wiedzy o człowieku w całym bogactwie jego słabości i niezwykłości. Tekst stanowi próbę odpowiedzi na pytanie o fenomen wiedeńskiego malarza, który z odwagą podążył za melancholijnymi tropami w jednym z przełomowych momentów historycznych – upadającej epoki
Classifier Based Graph Construction for Video Segmentation
No full textVideo segmentation has become an important and active research area with a large diversity of pro-posed approaches. Graph-based methods, enabling top-performance on recent benchmarks, consist of three essen-tial components: 1. powerful features account for object ap-pearance and motion similarities; 2. spatio-temporal neigh-borhoods of pixels or superpixels (the graph edges) are modeled using a combination of those features; 3. video segmentation is formulated as a graph partitioning prob-lem. While a wide variety of features have been explored and various graph partition algorithms have been pro-posed, there is surprisingly little research on how to con-struct a graph to obtain the best video segmentation perfor-mance. This is the focus of our paper. We propose to com-bine features by means of a classifier, use calibrated classi-fier outputs as edge weights and define the graph topology by edge selection. By learning the graph (without changes to the graph partitioning method), we improve the results of the best performing video segmentation algorithm by 6% on the challenging VSB100 benchmark, while reducing its runtime by 55%, as the learnt graph is much sparser
Spectral Graph Reduction for Efficient Image and Streaming Video Segmentation
No full textComputational and memory costs restrict spectral tech-niques to rather small graphs, which is a serious limitation especially in video segmentation. In this paper, we propose the use of a reduced graph based on superpixels. In con-trast to previous work, the reduced graph is reweighted such that the resulting segmentation is equivalent, under certain assumptions, to that of the full graph. We consider equiva-lence in terms of the normalized cut and of its spectral clus-tering relaxation. The proposed method reduces runtime and memory consumption and yields on par results in im-age and video segmentation. Further, it enables an efficient data representation and update for a new streaming video segmentation approach that also achieves state-of-the-art performance. 1
Learning must-link constraints for video segmentation based on spectral clustering
No full textIn recent years it has been shown that clustering and segmentation methods can greatly benefit from the integration of prior information in terms of must-link constraints. Very recently the use of such constraints has been integrated in a rigorous manner also in graph-based methods such as normalized cut. On the other hand spectral clustering as relaxation of the normalized cut has been shown to be among the best methods for video segmentation. In this paper we merge these two developments and propose to learn must-link constraints for video segmentation with spectral clustering. We show that the integration of learned must-link constraints not only improves the segmentation result but also significantly reduces the required runtime, making the use of costly spectral methods possible for today’s high quality video
The Development of an Efficient Grasp Master for Space Robotics Teleoperation
No full textThis thesis reports on the development of an efficient grasp master for space robotics teleoperation that will complement the haptic human arm master X-Arm-2 (Schiele 2011). It should provide the hand and fingers of various operators with sensing and feedback functions, enabling bilateral teleoperation of various types of robotic end-effectors in different control architectures to perform a wide range of tasks. Analysis of the state of the art of reported grasp masters reveals issues that limit their use in combination with arm master devices. Two important factors involved are device placement w.r.t. the operator (e.g. arm coverage by actuators and arm workspace limitation by external transmissions) and structural complexity (bulky designs leading to uncomfortable operation and mechanical losses). Because of these issues, the device efficiency, defined as the user- and device performance achieved with respect to the resources expended, appears to be generally too low. It is the goal of this work to present on the development of an efficiency grasp master device that can be used by various operators in space robotics teleoperation without requiring device adjustments. By following a human-centric design approach, considering relevant space operation tasks, required operator grasp types, and psychophysical effects in human grasping, a reduction of the required number of DOFs of a possible master device was achieved. In combination with the separation of control and feedback channels this can constitute an efficient device concept and was elaborated into a detailed prototype design in this work. As a tool in the design of device geometry and workspace verification, an adaptable kinematic human hand model was constructed and partially verified. This model is based on human functional anatomy and is easily adaptable to various real human hand sizes by the use of body proportions. Verification of the grasp master workspace showed robustness against hand size variation from the 5th female- till 95th male percentile. Verification by analysis and simulation showed that key human factors- and performance requirements can be met. Verification using device efficiency indicators, proposed for the purpose, indicates that the grasp master design is relatively efficient w.r.t. other compared devices based on slave controllability, slave observability, and the quality of the reflected force. The proposed prototype design will be manufactured to enable further verification by testing and validation of real user interaction.BMDBiomeMechanical, Maritime and Materials Engineerin
Improved Image Boundaries for Better Video Segmentation
No full textGraph-based video segmentation methods rely on superpixels as starting point. While most previous work has focused on the construction of the graph edges and weights as well as solving the graph partitioning problem, this paper focuses on better superpixels for video segmentation. We demonstrate by a comparative analysis that superpixels extracted from boundaries perform best, and show that boundary estimation can be significantly improved via image and time domain cues. With superpixels generated from our better boundaries we observe consistent improvement for two video segmentation methods in two different datasets
First International Workshop on Video Segmentation- Panel Discussion
No full textAbstract. Interest in video segmentation has grown significantly in re-cent years, resulting in a large body of works along with advances in both methods and datasets. Progress in video segmentation would enable new approaches to building 3D object models from video, understanding dy-namic scenes, robot-object interaction and several other high-level vision tasks. The workshop brought together a broad and representative group of video segmentation researchers working on a wide range of topics. This paper summarizes the panel discussion at the workshop, which focused on three questions: (1) Why does video segmentation currently not meet the performance of image segmentation and what difficulties prevent it from leveraging motion? (2) Is video segmentation a stand-alone prob-lem or should it rather be addressed in combination with recognition and reconstruction? (3) Which are the right video segmentation subtasks the field should focus on, and how can we measure progress
- …
