Embedded visual assist systems are emerging as increasingly viable tools for aiding visually impaired persons in their day-to-day life activities. Novel wearable devices with imaging capabilities will be uniquely posi...
详细信息
ISBN:
(纸本)9781479962785
Embedded visual assist systems are emerging as increasingly viable tools for aiding visually impaired persons in their day-to-day life activities. Novel wearable devices with imaging capabilities will be uniquely positioned to assist visually impaired in activities such as grocery shopping. However, supporting such time-sensitive applications on embedded platforms requires an intelligent trade-off between accuracy and computational efficiency. In order to maximize their utility in real-world scenarios, visual classifiers often need to recognize objects within large sets of object classes that are both diverse and deep. In a grocery market, simultaneously recognizing the appearance of people, shopping carts, and pasta is an example of a common diverse object classification task. Moreover, a useful visual-aid system would need deep classification capability to distinguish among the many styles and brands of pasta to direct attention to a particular box. Exemplar Support Vector Machines (ESVMs) provide a means of achieving this specificity, but are resource intensive as computation increases rapidly with the number of classes to be recognized. To maintain scalability without sacrificing accuracy, we examine the use of a biologically-inspired classifier (HMAX) as a front-end filter that can narrow the set of ESVMs to be evaluated. We show that a hierarchical classifier combining HMAX and ESVM performs better than either of the two individually. We achieve 12% improvement in accuracy over HMAX and 4% improvement over ESVM while reducing computational overhead of evaluating all possible exemplars.
This paper describes a new technique for recognizing speech using visual speech information. The video data of the speaker's mouth is represented using grayscale images named as motion history image (MHI). MHI is ...
详细信息
The characterization and abstraction of large multivariate time series data often poses challenges with respect to effectiveness or efficiency. Using the example of human motion capture data challenges exist in creati...
详细信息
The process of describing the distribution of impurities is described using two examples: phenomena in the bed of the Volga River in the northern section of the Volga-Akhtuba floodplain without a source under the infl...
详细信息
We present a method that allows for reconstructing non-stationary, time-varying gas flows around moving objects. Our work extends the background oriented Schlieren tomography (3D-BOS) acquisition technique to capture ...
详细信息
ISBN:
(纸本)9783980487481
We present a method that allows for reconstructing non-stationary, time-varying gas flows around moving objects. Our work extends the background oriented Schlieren tomography (3D-BOS) acquisition technique to capture gas flows also in the presence of occluding objects. An algorithm is presented that exploits the unique properties of BOS background patterns to robustly segment occluding objects. Numerical issues in the refractive index field reconstruction are addressed and successfully solved by the new method.
The medical domain provides excellent opportunities for the application of computergraphics, visualization, and virtual environments, with the potential to help improve healthcare and bring benefits to patients. Poss...
详细信息
The medical domain provides excellent opportunities for the application of computergraphics, visualization, and virtual environments, with the potential to help improve healthcare and bring benefits to patients. Possible applications include anatomical educational tools; patient education; diagnostic aids; virtual endoscopy; planning aids; guidance aids; skills training; and computer augmented reality. This talk provides a comprehensive overview of the state-of-the-art in this exciting field, including examples from research projects and commercially available products. The term cybermedicine was discussed and used to categorise those medical applications that can be delivered via the World Wide Web, preferably in the context of a collaborative virtual environment. The issues for effective cybermedicine was highlighted, and we will look ahead to future developments
We compare Growing Cell Structures and Growing Neural Gas, which were introduced by Bernd Fritzke and which are famous for their facilities in classification, clustering, dimensionality reduction, data visualization, ...
详细信息
ISBN:
(纸本)9789898565334
We compare Growing Cell Structures and Growing Neural Gas, which were introduced by Bernd Fritzke and which are famous for their facilities in classification, clustering, dimensionality reduction, data visualization, and approximation tasks. We practically test and analyze their capabilities in geometric approximation and focusing on the application of surface reconstruction from 3D point-data. Our focus is to work out the differences of the algorithms that are especially relevant concerning approximation purposes. We address the issue of suitable input data, their applied graphs, their topological properties, their run time complexities and we present a summary of suggested alternations to both approaches and evaluate our results.
Synthesis of hair images is one of the most important and challenging computergraphics problems. We propose a new technique for automatically generating realistic human hair texture on 3D models of human characters. ...
详细信息
Synthesis of hair images is one of the most important and challenging computergraphics problems. We propose a new technique for automatically generating realistic human hair texture on 3D models of human characters. The idea is inspired by the similarity between the texture of human hair and the texture generated by the LIC algorithm. The proposed technique generates the texture of human hair using the vector field defining the directions of hair strands as the input to the 3D LIC algorithm.
3D realtime modeling places a heavy load on CPU. This paper presents a new method on 3D visualization in reservoir modeling system by using the computation power of modern programmable graphics hardware (GPU). The pro...
详细信息
ISBN:
(纸本)9783662491553;9783662491546
3D realtime modeling places a heavy load on CPU. This paper presents a new method on 3D visualization in reservoir modeling system by using the computation power of modern programmable graphics hardware (GPU). The proposed scheme is devised to achieve parallel processing of massive reservoir logging data. By taking advantage of the GPU's parallel processing capability, moreover, the performance of our scheme is discussed in comparison with that of the implementation entirely running on CPU. Experimental results clearly show that the proposed parallel processing can remarkably accelerate the data clustering task. Especially, although data-transferring from GPU to CPU is generally costly, acceleration by GPU is significant to save the total execution time of data-clustering, and significantly alleviates the computing load on CPU.
computergraphics in combination with mobile devices find much use in the fields of entertainment, education and data displaying. The amount of information that is possible to provide to the user depends greatly on th...
详细信息
ISBN:
(纸本)9783319243061;9783319243054
computergraphics in combination with mobile devices find much use in the fields of entertainment, education and data displaying. The amount of information that is possible to provide to the user depends greatly on the optimization of graphic chain in the development of given application. There is a large area of web based software solutions where the speed and fluency of visualization and rendering are the most important parameters of user satisfaction when using of mobile client application. The important element is simplification of the scene by removing objects that are not currently visible. The methods that can be used as a solution for these problems are frustum culling, occlusion culling and level of detail. The work describes implementation of the first of mentioned methods on the Android platform and it includes a comparison of results from different devices when displaying a complicated 3D scene.
暂无评论