| 5631. |
- Kollreider, Klaus, et al.
(författare)
-
Real-Time Face Detection and Motion Analysis With Application in “Liveness” Assessment
- 2007
-
Ingår i: IEEE Transactions on Information Forensics and Security. - New York : IEEE Press. - 1556-6013 .- 1556-6021. ; 2:3 part 2, s. 548-558
-
Tidskriftsartikel (refereegranskat)abstract
- A robust face detection technique along with mouth localization, processing every frame in real time (video rate), is presented. Moreover, it is exploited for motion analysis onsite to verify "liveness" as well as to achieve lip reading of digits. A methodological novelty is the suggested quantized angle features ("quangles") being designed for illumination invariance without the need for preprocessing (e.g., histogram equalization). This is achieved by using both the gradient direction and the double angle direction (the structure tensor angle), and by ignoring the magnitude of the gradient. Boosting techniques are applied in a quantized feature space. A major benefit is reduced processing time (i.e., that the training of effective cascaded classifiers is feasible in very short time, less than 1 h for data sets of order 104). Scale invariance is implemented through the use of an image scale pyramid. We propose "liveness" verification barriers as applications for which a significant amount of computation is avoided when estimating motion. Novel strategies to avert advanced spoofing attempts (e.g., replayed videos which include person utterances) are demonstrated. We present favorable results on face detection for the YALE face test set and competitive results for the CMU-MIT frontal face test set as well as on "liveness" verification barriers.
|
|
| 5632. |
- Kollreider, Klaus, et al.
(författare)
-
Real-Time Face Detection Using Illumination Invariant Features
- 2007
-
Ingår i: Image Analysis. - Berlin : Springer. - 9783540730392 ; , s. 41-50
-
Konferensbidrag (refereegranskat)abstract
- A robust object/face detection technique processing every frame in real-time (video-rate) is presented. A methodological novelty are the suggested quantized angle features (“quangles”), being designed for illumination invariance without the need for pre-processing, e.g. histogram equalization. This is achieved by using both the gradient direction and the double angle direction (the structure tensor angle), and by ignoring the magnitude of the gradient. Boosting techniques are applied in a quantized feature space. Separable filtering and the use of lookup tables favor the detection speed. Furthermore, the gradient may then be reused for other tasks as well. A side effect is that the training of effective cascaded classifiers is feasible in very short time, less than 1 hour for data sets of order 104. We present favorable results on face detection, for several public databases (e.g. 93% Detection Rate at 1×10− 6 False Positive Rate on the CMU-MIT frontal face test set).
|
|
| 5633. |
- Kollreider, Klaus, et al.
(författare)
-
Verifying Liveness by Multiple Experts in Face Biometrics
- 2008
-
Ingår i: IEEE Conference on Computer Vision and Pattern Recognition Workshops. - New York : IEEE Press. - 9781424423392 - 9781424423408 ; , s. 1200-1205
-
Konferensbidrag (refereegranskat)abstract
- Resisting spoofing attempts via photographs and video playbacks is a vital issue for the success of face biometrics. Yet, the “liveness ” topic has only been partially studied in the past. In this paper we are suggesting a holistic liveness detection paradigm that collaborates with standard techniques in 2D face biometrics. The experiments show that many attacks are avertible via a combination of antispoofing measures. We have investigated the topic using real-time techniques and applied them to real-life spoofing scenarios in an indoor, yet uncontrolled environment.
|
|
| 5634. |
- Kolossov, Eugen, et al.
(författare)
-
Engraftment of engineered ES cell-derived cardiomyocytes but not BM cells restores contractile function to the infarcted myocardium
- 2006
-
Ingår i: Journal of Experimental Medicine. - New York, USA : Rockefeller University Press. - 0022-1007 .- 1540-9538. ; 203:10, s. 2315-2327
-
Tidskriftsartikel (refereegranskat)abstract
- Cellular cardiomyoplasty is an attractive option for the treatment of severe heart failure. It is, however, still unclear and controversial which is the most promising cell source. Therefore, we investigated and examined the fate and functional impact of bone marrow (BM) cells and embryonic stem cell (ES cell)-derived cardiomyocytes after transplantation into the infarcted mouse heart. This proved particularly challenging for the ES cells, as their enrichment into cardiomyocytes and their long-term engraftment and tumorigenicity are still poorly understood. We generated transgenic ES cells expressing puromycin resistance and enhanced green fluorescent protein cassettes under control of a cardiac-specific promoter. Puromycin selection resulted in a highly purified (>99%) cardiomyocyte population, and the yield of cardiomyocytes increased 6-10-fold because of induction of proliferation on purification. Long-term engraftment (4-5 months) was observed when co-transplanting selected ES cell-derived cardiomyocytes and fibroblasts into the injured heart of syngeneic mice, and no teratoma formation was found (n = 60). Although transplantation of ES cell-derived cardiomyocytes improved heart function, BM cells had no positive effects. Furthermore, no contribution of BM cells to cardiac, endothelial, or smooth muscle neogenesis was detected. Hence, our results demonstrate that ES-based cell therapy is a promising approach for the treatment of impaired myocardial function and provides better results than BM-derived cells.
|
|
| 5635. |
- Kolseth, Anna-Karin, et al.
(författare)
-
Influence of genetically modified organisms on agro-ecosystem processes
- 2015
-
Ingår i: Agriculture, Ecosystems & Environment. - Amsterdam : Elsevier BV. - 1873-2305 .- 0167-8809. ; 214, s. 96-106
-
Tidskriftsartikel (refereegranskat)abstract
- Biotechnology offers extensive possibilities to incorporate new traits into organisms. Genetically modified (GM) traits relevant for agro-ecosystems include traits such as pest resistance and herbicide tolerance in crop plants, increased growth rate in fish and livestock, and enhanced nitrogen-fixation capabilities of soil microbes. In this review, we evaluated the direct and indirect trait-specific effects of GM plants, microbes, and animals on ecosystem processes and found that most of the effects of genetically modified organisms (GMOs) on ecosystem processes are indirect and are the result of associated changes in management strategy rather than a direct effect of the GMOs. Conflicting results on the performance and effects of GMOs are frequently reported, especially regarding crop yield and impacts on soil organisms. This is partly because methods with different levels of resolution have been used in different ecological contexts. Overall, there is little evidence that the effects of GM traits on ecosystem processes act with different mechanisms from those of traits modified using conventional methods. However, little is known about trait-specific effects of GMOs on ecosystem processes even though GMOs have been used for more than three decades. In particular, studies linking genetically modified traits to ecosystem processes at longer time scales are rare, but needed for evaluating trait effects, especially in an evolutionary context. In addition, biotechnology may provide a unique tool for gaining insights into the links between traits and ecosystem processes when integrated into basic ecological research. (C) 2015 Elsevier B.V. All rights reserved.
|
|
| 5636. |
|
|
| 5637. |
- Konečný, Michal, et al.
(författare)
-
Enclosing the behavior of a hybrid automaton up to and beyond a Zeno point
- 2016
-
Ingår i: Nonlinear Analysis. - London : Elsevier. - 1751-570X .- 1878-7460. ; 20, s. 1-20
-
Tidskriftsartikel (refereegranskat)abstract
- Even simple hybrid automata like the classic bouncing ball can exhibit Zeno behavior. The existence of this type of behavior has so far forced a large class of simulators to either ignore some events or risk looping indefinitely. This in turn forces modelers to either insert ad-hoc restrictions to circumvent Zeno behavior or to abandon hybrid automata. To address this problem, we take a fresh look at event detection and localization. A key insight that emerges from this investigation is that an enclosure for a given time interval can be valid independent of the occurrence of a given event. Such an event can then even occur an unbounded number of times. This insight makes it possible to handle some types of Zeno behavior. If the post-Zeno state is defined explicitly in the given model of the hybrid automaton, the computed enclosure covers the corresponding trajectory that starts from the Zeno point through a restarted evolution. ©2015 The Authors. Published by Elsevier Ltd.
|
|
| 5638. |
- Konečný, Michal, et al.
(författare)
-
Enclosing the Behavior of a Hybrid System up to and Beyond a Zeno Point
- 2013
-
Ingår i: 2013 IEEE 1st International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2013. - Piscataway, N.J. : IEEE Press. - 9781479907984 - 9781479907977 - 9781479907960 ; , s. 120-125
-
Konferensbidrag (refereegranskat)abstract
- Even simple hybrid systems like the classic bouncing ball can exhibit Zeno behaviors. The existence of this type ofbehavior has so far forced simulators to either ignore some events or risk looping indefinitely. This in turn forces modelers to either insert ad hoc restrictions to circumvent Zeno behavior or to abandon hybrid modeling. To address this problem, we take a fresh look at event detection and localization. A key insight that emerges from this investigation is that an enclosure for a given time interval can be valid independently of the occurrence of agiven event. Such an event can then even occur an unbounded number of times, thus making it possible to handle certain types of Zeno behavior. © 2013 IEEE.
|
|
| 5639. |
|
|
| 5640. |
- Kontrimas, Vilius, et al.
(författare)
-
Neural networks based screening of real estate transactions
- 2007
-
Ingår i: Neural Network World. - Prague : Institute of Information and Computer Technology ASCR; Faculty of Transport, Czech Polytechnic University. - 1210-0552. ; 17:1, s. 17-30
-
Tidskriftsartikel (refereegranskat)abstract
- Aiming to hide the real money gains and to avoid taxes, fictive prices are sometimes recorded in the real estate transactions. This paper is concerned with artificial neural networks based screening of real estate transactions aiming to categorize them into "clear" and "fictitious" classes. The problem is treated as an outlier detection task. Both unsupervised and supervised approaches to outlier detection are studied here. The soft minimal hyper-sphere support vector machine (SVM) based novelty detector is employed to solve the task without the supervision. In the supervised case, the effectiveness of SVM, multilayer perceptron (MLP), and a committee based classification of the real estate transactions are studied. To give the user a deeper insight into the decisions provided by the models, the real estate transactions are not only categorized into "clear" and "fictitious" classes, but also mapped onto the self organizing map (SOM), where the regions of "clear", "doubtful" and "fictitious" transactions are identified. We demonstrate that the stability of the regions evolved in the SOM during training is rather high. The experimental investigations performed on two real data sets have shown that the categorization accuracy obtained from the supervised approaches is considerably higher than that obtained from the unsupervised one. The obtained accuracy is high enough for the technique to be used in practice. © ICS AS CR 2007.
|
|
