May 2013
| No | Date | Title | Speaker | Presentation | Discussion |
| 1 | 2013-05-14 | A Compressive Sensing and Unmixing Scheme for Hyperspectral Data Processing | Woongbi | (PDF) | |
| In this paper, a low complexity scheme is proposed for hyperspectral data compression and reconstruction. The data reconstruction minimizes the total variation of the abundance fractions subject to a preprocessed fidelity equation with a significantly reduced size, and other side constraints. | |||||
| 2 | 2013- 05-21 | Dry and NOncontact EEG sensors for Mobile Brain-Computer Interfaces | Seungchan | (pdf) | |
| Dry and noncontact electroencephalographic (EEG) electrodes, which do not require gel or even direct scalp coupling, have been considered as an enabler of practical, real-world, brain–computer interface (BCI) platforms. This study compares wet electrodes to dry and through hair, noncontact electrodes within a steady state visual evoked potential (SSVEP) BCI paradigm. The construction of a dry contact electrode, featuring fingered contact posts and active buffering circuitry is presented. Additionally, the development of a new, noncontact, capacitive electrode that utilizes a custom integrated, high-impedance analog front-end is introduced. | |||||
| 3 | 2013-05-28 | Exemplar-Based Processing for Speech Recognition | Pavel | (PDF) |
|
| Automatic speech recognition has been dominated by technique using hidden Markov models to model time varying aspects of the acoustics. This method allows for a generalization of the observed data as long as the distribution estimated by the model is a reasonable description of the unseen data. In many cases, such a description must be simplified to allow reliable estimates of all free parameters in the model, and as a result, fine details in the model are lost. Exemplar based models have the potential to address this deficiency by building an instance of model based only on the relevant exemplars selected, exploiting sparse coding and compressive sensing technique. | |||||