Recognizing human actions in video sequences is frequently based on analyzing the shape of the human silhouette as the main feature. In this paper we introduce a method for recognizing different actions by comparing s...
详细信息
We present an approach for unsupervised alignment of an ensemble of images called congealing. Our algorithm is based on image registration using the mutual information measure as a cost function. The cost function is ...
详细信息
Modelling and reconstruction of tubular objects from graph representations is frequently needed in medical visualization. The constructed geometrical models need to be consistent and compact at the same time for inter...
详细信息
ISBN:
(数字)9783540264316
ISBN:
(纸本)3540250522
Modelling and reconstruction of tubular objects from graph representations is frequently needed in medical visualization. The constructed geometrical models need to be consistent and compact at the same time for interactive applications. In this paper we present a new method for generating compact topologically consistent 2-manifold surfaces of branching tubular objects. The proposed two-stage approach based on deformable simplex-meshes can handle higher order furcations.
This paper presents a new approach for constructing a consistent (non-manifold) mesh based on non-binary datasets. Mesh generation is the method of choice if a surface is extracted based on labeled volumetric datasets...
详细信息
ISBN:
(数字)9783540264316
ISBN:
(纸本)3540250522
This paper presents a new approach for constructing a consistent (non-manifold) mesh based on non-binary datasets. Mesh generation is the method of choice if a surface is extracted based on labeled volumetric datasets. In most cases, the traditional Marching Cubes algorithm or one of its extensions is applied for generating meshes of binary labeled input data. However, for different - especially medical - applications, non-binary datasets with more than two differently labeled voxels can be adjacent in a rectilinear grid. For these cases, the traditional Marching Cubes algorithm fails. Additional to the mesh generation algorithm, an extension for a well-known geometric filter is presented avoiding ridges in the mesh.
Current Internet-inspired mapping data are in the form of street maps, orthophotos, 3D models or street-side images to support searches and navigation. Images represent predominantly "eye candy" with little ...
详细信息
We introduce WebCAME, a client/server multiresolution rendering system for progressive transmission and visualization of compressed non-manifold triangle meshes with texture and color. The tool is implemented as a web...
详细信息
We introduce WebCAME, a client/server multiresolution rendering system for progressive transmission and visualization of compressed non-manifold triangle meshes with texture and color. The tool is implemented as a web browser plugin built upon standard components such as Qt, OpenGL, and ODBC. By utilizing and extending recently developed multiresolution techniques, it can provide view-dependent access to huge 3D data sets. With a size of less than 250kB it is small enough to be downloaded and installed within reasonable time. WebCAME comes with a preliminary GUI for navigation and parameter selection for the multiresolution renderer. CAME stands for Compressed Adaptive Multiresolution Encoding.
Accurate and realistic 3-dimensional models of urban environments are increasingly important for applications like virtual tourism, city planning, internet search and many emerging opportunities in the context of &quo...
详细信息
ISBN:
(纸本)9781617389160
Accurate and realistic 3-dimensional models of urban environments are increasingly important for applications like virtual tourism, city planning, internet search and many emerging opportunities in the context of "ambient intelligence". Applications like Bing-Maps or Google Earth are offering virtual models of many major urban areas worldwide. While initially, these data sets support visualization they are inherently capable of addressing a broader purpose. On the horizon are urban models that consist of semantically interpreted objects;an urban 3D visualization will be computer generated, with a fundamental advantage: the urban models can be searched based on object classes. This paper presents a framework which specifies the processing steps that are necessary for a reasonable semantic interpretation and evaluation of real property using high resolution aerial images. We first describe the different source data which have to be brought into a common coordinate system. In this process we build an integrated geometry and semantic object data set that can be analyzed for various purposes. Our focus is on characterizing individual properties and to determine the size of buildings, their number of floors, status of vegetation, roof shapes with chimneys and sky lights etc. We start out by merging the aerial imagery with the cadastral information to define each property as a separate entity for further analysis. The cadastral data may also contain preliminary information about a building footprint. In the next step the building footprints get refined vis-à-vis the mere cadastral prediction, using an image classification and the definition of roof lines. 3-D façade coordinates are computed from aerial image segments, the cadastral information and the DTM. This helps to determine the number of floors and the window locations to see if there are attics and basement windows. The paper uses an experimental data set of the City of graz (Austria) with 216 buildings representing
In this paper, we present an efficient solution for automatic detection and reading of dangerous goods plates on trucks and trains. According to the ADR agreement dangerous goods transports are marked with an orange p...
详细信息
The fields of Wearable Computing, Augmented Reality and Ubiquitous Computing are in principle highly convergent, as they all promise a utopian future in which the devices embedded in the environment, our bodies and ou...
详细信息
In this chapter we will introduce a new effective framework for learning an object detector. The main idea is to minimize the manual effort when learning a classifier and to combine the power of a discriminative class...
详细信息
In this chapter we will introduce a new effective framework for learning an object detector. The main idea is to minimize the manual effort when learning a classifier and to combine the power of a discriminative classifier with the robustness of a generative model. Starting with motion detection an initial set of positive examples is obtained by analyzing the geometry (aspect ratio) of the motion blobs. Using these samples a discriminative classifier is trained using an online version of AdaBoost. In fact, applying this classifier nearly all objects are detected but there is a great number of false positives. Thus, we apply a generative classifier to verify the obtained detections and to decide if a detected patch represents the object of interest or not. As we have a huge amount of data (video stream) we can be very conservative and use only patches for (positive or negative) updates if we are very confident about our decision. Applying this update rules, an incrementally better classifier is obtained without any user interaction. Moreover, an already trained classifier can be retrained online and can therefore easily be adapted to a completely different scene. We demonstrate the framework on different scenarios including pedestrian and car detection.
暂无评论