Software libraries and datasets that have been developed and/or collected by the SPL team are available at our Github repositories. If you find any of this material useful for your research, please give credit in your publications where it is due.
|Software Library||Description||Research Area||Language / Environment|
|FORTH TRACE LIBRARY||FORTH-TRACE library, as part of the FORTH-TRACE benchmark framework, is a Java implemented software tool that covers the entire feature-level fusion chain applied in Human Activity Recognition (HAR) applications.||Distributed Sensor Networks||JAVA|
|Feature selection & performance characterization of multi-hop WSN||This repository contains MATLAB code for the Representation Entropy Clustering Feature Selection Algorithm, which uses representation entropy ranking & clustering for the automatic calculation of the dominant network features that characterize the network performance in multi-hop wireless sensor network deployments. A sample dataset is also included.||Distributed Sensor Networks||MATLAB|
|Spectral Super Resolution of Hyperspectral Images||This repository contains MATLAB codes and scripts designed for the spectral super-resolution of hyperspectral data. The proposed approach synthesizes a high spectral resolution 3D data cube from its acquired low resolution form, by capitalizing on the Sparse Representations (SR) learning framework.||Imaging||MATLAB|
|Denoising||This repository contains MATLAB codes and scripts designed for the denoising of spectroscopic data.||Imaging||MATLAB|
|1-bit Tensor Completion||This repository contains MATLAB codes and scripts designed for the 1-bit tensor completion algorithm. In the proposed approach is explored the recovery of a low-rank tensor from a small number of binary measurements. Specifically, given a 3-order tensor where only a small number of binary entries are available, we unfold it into 3 matrices and we apply the quantized matrix completion algorithm to the all-mode matricizations of the tensor.||Imaging||MATLAB|
|Tensor Dictionary Learning with representation quantization for Remote Sensing Observation Compression||This repository contains MATLAB codes and scrips designed for the compression of tensor data based on a novel tensor dictionary learning method that uses the CANDECOMP/PARAFAC (CP) decomposition, as it is presented in the paper "Tensor Dictionary Learning with representation quantization for Remote Sensing Observation Compression" (A. Aidini, G. Tsagkatakis, P. Tsakalides). In the proposed method, a dictionary of specially structured tensors is estimated using an Alternating Direction Method of Multipliers (ADMM) approach, as well a symbol encoding dictionary is learned from training samples. Given the learned models, a new sample is first presented by a set of sparse coefficients corresponding to linear combination of the elements of the learned dictionary. Then, the derived coefficients are quantized and encoded in order to be transmitted, significantly reducing the number of bits required to represent the useful information of the data.||Imaging||MATLAB|
|Multi-label Land Cover Scene Categorization||This repository contains Python code designed for the the problem of multi-label land cover scene categorization. A deep learning architecture is adopted, namely Convolutional Neural Networks, along with the utilization of a data augmentation technique for the artificial increase of the size of the employed dataset.||Imaging||PYTHON|
|Dataset||Description||Research Area||File Size|
|The FORTH TRACE Dataset||This is a Human Activity Recognition dataset, which contains 3-axial sensing parameters (acceleration, angular velocity, magnetic field variations) from five different body locations. 15 healthy volunteers participated this study, performing a set of 16 activities.||Distributed Sensor Networks||108MB|
|Microphone array recordings for localization in real-life conditions||Speech recordings with an acoustic sensor network of two microphone arrays in a typical office environment (T60 approximately equal to 400 ms).||Acoustic Sensor Networks||75MB|
|User generated audio recordings||Audio recordings captured using smartphones and portable electronic devices at two different public events, a musical concert and a football match.||Audio & Speech Signal Processing|
Copyright (c) 2011-2016, Signal Processing Lab (SPL), Institute of Computer Science (ICS), FORTH, Greece.
All rights reserved.
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
3. Neither the name of the Signal Processing Lab, the Institute of Computer Science and FORTH, nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
THE SOFTWARE LIBRARIES AND DATASETS ARE PROVIDED BY THE INSTITUTE AND CONTRIBUTORS “AS IS” AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.