Computer Vision – ACCV 2009, 9th Asian Conference on Computer Vision, Xi’an, September 23-27, 2009, Revised Selected Papers, Part III

Human activity recognition using MotionNode sensors is getting prominence effect in our daily life logs. Providing accurate information on human's activities and behaviors is one of the most challenging tasks in ubiquitous computing and human-Computer interaction. In this paper, we proposed an efficient model for having statistical features along SMO-based random forest. Initially, we processed a 1-D Hadamard transform wavelet and 1-D LBP based extraction algorithm to extract valuable features. For activity classification, we used sequential minimal optimization along with Random Forest over two benchmarks USC-HAD dataset and IMSB datasets. Experimental results show that our proposed model can compete with other state-of-the-art methods and can be effectively used to recognize robust human activities in terms of efficiency and accuracy.

Wearable Sensors for Activity Analysis using SMO-based Random Forest over Smart home and Sports Datasets

Clustering with similarity preserving

Odor-induced emotion recognition based on average frequency band division of EEG signals

Objective Convolutional neural networks (CNNs) have proven successful as function approximators and have therefore been used for classification problems including electroencephalography (EEG) signal decoding for brain-computer interfaces (BCI). Artificial neural networks, however, are considered black boxes, because they usually have thousands of parameters, making interpretation of their internal processes challenging. Here we systematically evaluate the use of CNNs for EEG signal decoding and investigate a method for visualizing the CNN model decision process. Approach We developed a CNN model to decode the covert focus of attention from EEG event-related potentials during object selection. We compared the CNN and the commonly used linear discriminant analysis (LDA) classifier performance, applied to datasets with different dimensionality, and analyzed transfer learning capacity. Moreover, we validated the impact of single model components by systematically altering the model. Furthermore, we investigated the use of saliency maps as a tool for visualizing the spatial and temporal features driving the model output. Main results The CNN model and the LDA classifier achieved comparable accuracy on the lower-dimensional dataset, but CNN exceeded LDA performance significantly on the higher-dimensional dataset (without hypothesis-driven preprocessing), achieving an average decoding accuracy of 90.7% (chance level = 8.3%). Parallel convolutions, tanh or ELU activation functions, and dropout regularization proved valuable for model performance, whereas the sequential convolutions, ReLU activation function, and batch normalization components reduced accuracy or yielded no significant difference. Saliency maps revealed meaningful features, displaying the typical spatial distribution and latency of the P300 component expected during this task. Significance Following systematic evaluation, we provide recommendations for when and how to use CNN models in EEG decoding. Moreover, we propose a new approach for investigating the neural correlates of a cognitive task by training CNN models on raw high-dimensional EEG data and utilizing saliency maps for relevant feature extraction.

https://iopscience.iop.org/article/10.1088/1741-2552/ab3bb4/pdf

Convolutional neural networks for decoding of covert attention focus and saliency maps for EEG feature visualization.

Spectral clustering techniques have shown their capability to identify the data relationships using graph analysis, achieving better accuracy than traditional algorithms as k-means. Here, we propose a methodology to build automatically a graph representation over the input data for spectral clustering based approaches by taking into account the local and global sample structure. Regarding this, both the Euclidean and the geodesic distances are used to identify the main relationships between a given point and neighboring samples around it. Then, given the information about the local data structure, we estimate an affinity matrix by means of Gaussian kernel. Synthetic and real-world datasets are tested. Attained results show how our approach outperforms, in most of the cases, benchmark methods.

/pdf/automatic-graph-building-approach-for-spectral-clustering-1hqslf1r8e.pdf

Automatic Graph Building Approach for Spectral Clustering

Sign language recognition is a highly-complex problem due to the amount of static and dynamic gestures needed to represent such language, especially when it changes from country to country. This article focuses on static recognition of vowels in Colombian Sign Language. A total of 151 images were acquired for each class, and an additional non-vowel class with different scenes was also considered. The object of interest was cut out of the rest of the scene in the captured image by using color information. Subsequently, features were extracted to describe the gesture that corresponds to a vowel or to the class that does not match any vowel. Next, four sets of features were selected. The first one contained all of them; from it, three new sets were generated. The second one was extracted from a subset of features given by the Principal Component Analysis (PCA) algorithm. The third set was obtained by Sequential Feature Selection (SFS) with the FISHER measure. The last set was completed with SFS based on the performance of the K-Nearest Neighbor (KNN) algorithm. Finally, multiple classifiers were tested on each set by cross-validation. Most of the classifiers achieved a performance over 90%, which led to conclude that the proposed method allows an appropriate class distinction.

/pdf/clasificacion-automatica-de-las-vocales-en-el-lenguaje-de-3xg9xl77tr.pdf

Clasificación automática de las vocales en el lenguaje de señas colombiano

This paper presents the development of a unified view of spectral clustering and unsupervised dimensionality reduction approaches within a generalized kernel framework. To do so, the authors propose a multipurpose latent variable model in terms of a high-dimensional representation of the input data matrix, which is incorporated into a least-squares support vector machine to yield a generalized optimization problem. After solving it via a primal-dual procedure, the final model results in a versatile projected matrix able to represent data in a low-dimensional space, as well as to provide information about clusters. Specifically, our formulation yields solutions for kernel spectral clustering and weighted-kernel principal component analysis.

On the relationship between dimensionality reduction and spectral clustering from a kernel viewpoint

Spectral clustering has shown to be a powerful technique for grouping and/or rank data as well as a proper alternative for unlabeled problems. Particularly, it is a suitable alternative when dealing with pattern recognition problems involving highly hardly separable classes. Due to its versatility, applicability and feasibility, this clustering technique results appealing for many applications. Nevertheless, conventional spectral clustering approaches lack the ability to process dynamic or time-varying data. Within a spectral framework, this work presents an overview of clustering techniques as well as their extensions to dynamic data analysis.

Andrés Eduardo Castro-Ospina

Papers

Automatic Graph Building Approach for Spectral Clustering

Clasificación automática de las vocales en el lenguaje de señas colombiano

On the relationship between dimensionality reduction and spectral clustering from a kernel viewpoint

On the Spectral Clustering for Dynamic Data

Characterization and classification of intracardiac atrial fibrillation signals using the time-singularity multifractal spectrum distribution