Alzheimer’s Disease (AD) and Frontotemporal Dementia (FTD), among the most prevalent neurodegenerative disorders, disrupt brain activity and connectivity, highlighting the need for tools that can effectively capture these alterations. Effective Connectivity Networks (ECNs), which model causal interactions between brain regions, offer a promising approach to characterizing AD and FTD related neural changes. In this study, we estimate ECNs from EEG traces using a state-of-the-art causal discovery method specifically designed for time-series data, to recover the causal structure of the interactions between brain areas. The recovered ECNs are integrated into a novel Graph Neural Network architecture (ECoGNet), where nodes represent brain regions and edge features encode causal relationships. Our method combines ECNs with features summarizing local brain dynamics to improve AD and FTD detection. Evaluated on a publicly available EEG dataset, the proposed approach demonstrates superior performance compared to models that either use non-causal connectivity networks or omit connectivity information entirely.

Facial Landmark Detection (FLD) algorithms play a crucial role in numerous computer vision applications, particularly in tasks such as face recognition, head pose estimation, and facial expression analysis. While FLD on images has long been the focus, the emergence of 3D data has led to a surge of interest in FLD on it due to its potential applications in various fields, including medical research. However, automating FLD in this context presents significant challenges, such as selecting suitable network architectures, refining outputs for precise landmark localization and optimizing computational efficiency. In response, this paper presents a novel approach, the 2-Stage Stratified Graph Convolutional Network ( 2S-SGCN ), which addresses these challenges comprehensively. The first stage aims to detect landmark regions using heatmap regression, which leverages both local and long-range dependencies through a stratified approach. In the second stage, 3D landmarks are precisely determined using a new post-processing technique, namely MSE-over-mesh . 2S-SGCN ensures both efficiency and suitability for resource-constrained devices. Experimental results on 3D scans from the public Facescape and Headspace datasets, as well as on point clouds derived from FLAME meshes collected in the DAD-3DHeads dataset, demonstrate that the proposed method achieves state-of-the-art performance across various conditions. Source code is accessible at https://github.com/gfacchi-dev/CVIU-2S-SGCN . • 2S-SGCN: a novel dual stage approach for 3D landmark detection. • Stratified GCN heatmap regressor. • MSE-over-mesh: FLD from the estimated heatmap using MSE on the original mesh. • State-of-the-art results on Facescape and Headspace datasets. • Evaluation of 2D-SGCN on DAD-3DH dataset for 3D FLD on 2D data.

This study examines seniors’ creative engagement in group activities using synchronous communication tools and explores automatic assessment methods through behavioral and psychophysiological measurements. Working with a small senior group on collaborative creative tasks, we implemented a comprehensive data collection approach using audio-visual and physiological measurements. Machine learning models were used to evaluate group creative engagement levels using various data subsets. Results show that engagement assessment can be effective with different feature combinations, allowing flexibility across contexts and constraints. The multimodal approach, combining facial, audio, and body analysis, achieved optimal performance and is recommended when conditions permit. Our research provides insights into seniors’ online creative participation and presents an automated system for detecting creative engagement in virtual teams, supporting active participation strategies.

This study investigates the potential of Graph Neural Networks (GNNs) for analyzing 3D facial morphology, leveraging facial landmarks as graph nodes to capture the intrinsic structure of 3D face scans. This research evaluates the effectiveness of three distinct approaches for defining graph vertices by associating them with: (1) a well-established set of anthropometric landmarks identified through tactile assessment, widely considered the gold standard in facial anthropometry; (2) automatically detected 3D facial keypoints estimated using advanced algorithms; and (3) geometry-based random point cloud sub-sampling via farthest point sampling (FPS). To evaluate the effectiveness of GNNs and facial landmarks in capturing and representing meaningful morphological patterns, the study employs two benchmark tasks: gender classification and age regression. Extensive experiments across various GNN architectures and three datasets — each presenting diverse and challenging conditions — demonstrate that semantically meaningful landmarks, whether anthropometric or automatically detected, consistently outperform non-semantic random samples in both tasks and across all datasets. These results highlight the crucial role of semantic contextualization in graph-based facial analysis. Notably, models utilizing automatically detected facial keypoints achieved performance comparable to those based on manually annotated anthropometric landmarks, offering a scalable and cost-effective alternative without compromising accuracy. These findings support the integration of automated GNN-based methodologies into a wide range of applications, including clinical diagnosis, forensic analysis, and biometric recognition. • Learning 3D facial features to classify gender and age. • Comparing anthropometric vs. automatic landmarks and FPS sub-sampling. • Characterizing graph nodes with local and global geometric features. • Comparing GCN, GAT, DynamicEdgeConv, and Transformer models.

We live in an era of privacy concerns, motivating a large research effort in face de-identification. As in other fields, we are observing a general movement from hand-crafted to deep learning methods, mainly involving generative models. Although these methods produce more natural de-identified images or videos, we claim that the mere evaluation of the de-identification is not sufficient, especially when it comes to processing the images/videos further. In this note, we take into account the issue of preserving privacy, facial expressions, and photo-reality simultaneously, proposing a general testing framework. The quantitative evaluation is applied to four open-source tools, producing a baseline for future de-identification methods.

We present AMHUSE (A Multimodal dataset for HUmour SEnsing) along with a novel web-based annotation tool named DANTE (Dimensional ANnotation Tool for Emotions). The dataset is the result of an experiment concerning amusement elicitation, involving 36 subjects in order to record the reactions in presence of 3 amusing and 1 neutral video stimuli. Gathered data include RGB video and depth sequences along with physiological responses (electrodermal activity, blood volume pulse, temperature). The videos were later annotated by 4 experts in terms of valence and arousal continuous dimensions. Both the dataset and the annotation tool are made publicly available for research purposes.

For decades, face recognition (FR) has attracted a lot of attention, and several systems have been successfully developed to solve this problem. However, the issue deserves further research effort so as to reduce the still existing gap between the computer and human ability in solving it. Among the others, one of the human skills concerns his ability in naturally conferring a “degree of reliability” to the face identification he carried out. We believe that providing a FR system with this feature would be of great help in real application contexts, making more flexible and treatable the identification process. In this spirit, we propose a completely automatic FR system robust to possible adverse illuminations and facial expression variations that provides together with the identity the corresponding degree of reliability. The method promotes sparse coding of multi-feature representations with LDA projections for dimensionality reduction, and uses a multistage classifier. The method has been evaluated in the challenging condition of having few (3–5) images per subject in the gallery. Extended experiments on several challenging databases (frontal faces of Extended YaleB, BANCA, FRGC v2.0, and frontal faces of Multi-PIE) show that our method outperforms several state-of-the-art sparse coding FR systems, thus demonstrating its effectiveness and generalizability.

In this paper we present a new local-based face recognition system that combines weak classifiers to create a robust system able to recognize faces in presence of either occlusions or large expression variations. The method relies on sparse approximation using dictionaries built on local features. Experiments on the AR database show the effectiveness of our method, which achieves better performance than those obtained by the state-of-the-art ℓ1norm-based sparse representation classifier (SRC).

The literature on the topic has shown a strong correlation between the degree of precision of face localization and the face recognition performance. Hence, there is a need for precise facial feature detectors, as well as objective measures for their evaluation and comparison. In this paper, we will present significant improvements to a previous method for precise eye center localization, by integrating a module for mouth localization. The technique is based on Support Vector Machines trained on optimally chosen Haar wavelet coefficients. The method has been tested on several public databases; the results are reported and compared according to a standard error measure. The tests show that the algorithm achieves high precision of localization.

We present a method for precise eye localization that uses two Support Vector Machines trained on properly selected Haar wavelet coefficients. The evaluation of our technique on many standard databases exhibits very good performance. Furthermore, we study the strong correlation between the eye localization error and the face recognition rate.

We describe an algorithm for the analysis of color images of basins which gather dirty water, with the objective of detecting and estimate the presence of exceeding combustible oil. The algorithm first detects most of the pixel corresponding to the combustible oil applying a thresholding method based on the entropy of the image's histogram. Then it treats the images differently according to the first order statistics of the image's hue; we distinguished two cases in which the algorithm has a different behavior: in the first case, it applies a clustering method in the HSV color space, whereas in the second case it. operates in the CIE-L*a*b* color space. Finally it estimates the percentage of the present oil.

We describe a three-step algorithm for the analysis of color images of flames, with the objective of analyzing the combustion process and its control parameters indirectly. The algorithm first extracts the interesting regions, applying a clustering method in the RGB color space, identifying and eliminating irrelevant regions and applying morphological operators. Then it calculates different geometrical parameters for each image. In the final step, it produces the temporal trend and the frequency spectrum for each parameter. The spectra are considered useful for the validation of a theoretical model about the instability phenomenon generated by premixed flames.