In this paper, we propose a novel local feature, called Local Orientation Adaptive Descriptor (LOAD), to capture regional texture in an image. In LOAD, we proposed to define point description on an Adaptive Coordinate System (ACS), adopt a binary sequence descriptor to capture relationships between one point and its neighbors and use multi-scale strategy to enhance the discriminative power of t...

Adaptive hypermedia courseware bene ts from being distributed over theWeb: content can always be kept up-to-date, discussions and interactions between instructors and learners can be supported, new courses can easily be distributed to the students. Nevertheless, adaptive hypermedia systems are even in the web content still stand-alone systems as long as they lack the ability to integrate and ad...

Learning Management Systems such as Blackboard, Moodle, Sakai and Claroline focus on supporting the learning process at a fairly global level of courses and tests. Recent additions focus on interaction (discussion forums, chat rooms and wikis). GRAPPLE tackles an important omission: integrating the adaptive delivery of course material into the supported learning process. GRAPPLE is an EU funded...

t=1 Xt ≤ 2 max { 2 √ σ, √ ln( 1δ ) }√ ln( 1δ ) = Õ( √ σ), by choosing δ 1 log2(T ) . This implies inequality (2). Inequality (3) is proved similarly. Note that these high probability bounds are conditioned on the internal randomness of WL. By taking an expectation of this conditional probability over the internal randomness of WL, we conclude that inequalities (2) and (3) hold with high probabi...

Figure 3 illustrates detection results. Our algorithm is able to detect objects under different scales, lighting conditions, or partial occlusions. Some failure cases are also shown in Figure 4, where most of them are caused by heavy occlusions or objects with similar appearance. Moreover, we present the region proposal results that generate the bounding boxes. It shows that our SVR models are ...

An intelligent approach for smart material actuator modeling of the actuation lines in a morphing wing system is presented, based on adaptive neuro-fuzzy inference systems. Four independent neuro-fuzzy controllers are created from the experimental data using a hybrid method -a combination of back-propagation and Least-Mean-Square (LMS) methods -to train the fuzzy inference systems. The controll...

This paper investigates memory carving techniques for embedded devices. Given that cryptographic material in memory dumps makes carving techniques inefficient, we introduce a methodology to distinguish meaningful information from cryptographic material in smallsized memory dumps. The proposed methodology uses an adaptive boosting technique with statistical tests. Experimented on EMV cards, the ...