196,070 research outputs found
A Range/Domain Approximation Error Based Approach for Fractal Image Compression
Get PDFFractals can be an effective approach for several applications
other than image coding and transmission: database indexing,
texture mapping, and even pattern recognition problems
such as writer authentication. However, fractal-based algorithms
are strongly asymmetric because, in spite of the linearity of the decoding
phase, the coding process is much more time consuming.
Many different solutions have been proposed for this problem, but
there is not yet a standard for fractal coding. This paper proposes
a method to reduce the complexity of the image coding phase by
classifying the blocks according to an approximation error measure.
It is formally shown that postponing range\slash domain comparisons
with respect to a preset block, it is possible to reduce
drastically the amount of operations needed to encode each range.
The proposed method has been compared with three other fractal
coding methods, showing under which circumstances it performs
better in terms of both bit rate and/or computing time
MUBIDUS-I: A multibiometric and multipurpose dataset
No full textIndividual biometric traits can seldom fulfill the requirements of security systems in the wild, so researchers were led to investigate multi-biometric/multi-modal systems. This has produced increasing demand for datasets suitable for validating multi-trait and multi-modal biometric systems. Recent devices available for image acquisition and processing can provide a wide range of data sources for biometric applications. The purpose of this work is to present a new multi-biometric dataset that includes a number of traits and acquisition devices wider than most existing datasets. It includes images and videos acquired from 80 subjects in an indoor and outdoor environment, in controlled and non-controlled conditions. Traits such as face, periocular regions, ear, iris, and others are acquired by cameras, mobile devices, and a drone. The data are structured to support experiments adhering to the most common protocols in the literature
Measuring Prefrontal Hemodynamic Responses Using Functional Near-Infrared Spectroscopy During Mobility for a Child With Motor Impairment
No full textAbstract
Date Presented 3/31/2017
This study used functional near-infrared spectroscopy (fNIRS) to identify changes in cognitive workload in a child with motor impairment during experiences with robot-assisted mobility. The study provides preliminary support for using fNIRS to measure cognitive workload in novel motor tasks.
Primary Author and Speaker: Kelly Cusick
Additional Authors and Speakers: Alexandra DiStasi, Stephanie Holowinski, Olivia Fitzpatrick
Contributing Authors: Megan Davis, Melody H. Wallace, Sharon A. Stansfield, Carole Dennis, Hélène M. Larin, Nancy Rader, Judith Pena-Shaff</jats:p
Face Authentication using Speed Fractal Technique
Get PDFIn this paper, a new fractal based recognition method, Face Authentication using Speed Fractal Technique (FAST), is presented. The main contribution is the good compromise between memory requirements, execution time and recognition ratio. FAST is based on Iterated Function Systems (IFS) theory, largely studied in still image compression and indexing, but not yet widely used for face recognition. Indeed, Fractals are well known to be invariant to a large set of global transformations. FAST is robust with respect to meaningful variations in facial expression and to the small changes of illumination and pose. Another advantage of the FAST strategy consists in the speed up that it introduces. The typical slowness of fractal image compression is avoided by exploiting only the indexing phase, which requires time O(D log (D)), where D is the size of the domain pool. Lastly, the FAST algorithm compares well to a large set of other recognition methods, as underlined in the experimental results
Ollivier-Ricci Curvature For Head Pose Estimation From a Single Image
No full textHead pose estimation is not only a crucial challenge for many real-world applications, such as driver attention detection analysis, but it represents an interesting strategy to support biometric frameworks as well. This paper aims to estimate head pose from a single image by applying notions of network curvature. In the real world, many complex networks have groups of nodes that are well connected to each other with significant functional roles. Similarly, the interactions of facial landmarks can be represented as complex dynamic systems modeled by weighted graphs. The functionality of such a system is therefore intrinsically linked to the topology and geometry of the underlying graph. In this work, using the geometric notion of Ollivier-Ricci curvature (ORC) on weighted graphs as input to the XGBoost regression model, we show that the intrinsic geometric basis of ORC offers a natural approach to discovering underlying common structure within a pool of poses. Experiments on the BIWI, AFLW2000 and Pointing`04 datasets show that the ORC XGB method performs well compared to state-ofthe-art methods, both landmark-based and image-only
- …
