Understanding how climate variability influences asphalt pavement surface temperatures is critical for assessing the durability and long-term performance of bituminous materials in road construction. This study develops a computational framework that integrates physical and data-driven modeling techniques to characterize asphalt surface temperature dynamics across diverse Australian climates. Using high-resolution climate records from over 450 meteorological stations, we established a comprehensive pipeline for data preprocessing, imputation, and hybrid feature selection through Self-Organizing Maps (SOM), Gaussian Mixture Models (GMM), and SHapley Additive exPlanations (SHAP), offering explainability and cluster-wise insights into regional climate behavior. To address spatial heterogeneity in thermal loading, a national climate zoning scheme was developed using k-means clustering, enabling targeted material performance assessment across climate zones. Within this framework, four predictive models were designed and evaluated: a climate-driven multiple linear regression model (C-MLR), an annual climatic pavement stress model (ACPS), a physically-informed heat flux model (H-FLUX), and a symbolic regression model based on linear genetic programming (LGP). Bayesian optimization and Limited-memory Broyden–Fletcher–Goldfarb–Shanno with Bounds (L-BFGS-B) tuning enhanced each model's predictive accuracy, with LGP outperforming others while maintaining model interpretability. The resulting high-resolution asphalt surface temperature maps inform the evaluation of oxidative aging, rutting, and thermal cracking risks. By linking climate-driven temperature variation to pavement distress mechanisms, this work provides a transferable framework for engineering design, offering insights for the development of climate-resilient road infrastructure. This approach demonstrates the effective use of Artificial Intelligence (AI) techniques for real-world civil engineering challenges under environmental uncertainty.

Abstract Growing application of artificial intelligence in geotechnical engineering has been observed; however, its ability to predict the properties and nonlinear behaviour of reactive soil is currently not well considered. Although previous studies provided linear correlations between shrink–swell index and Atterberg limits, obtained model accuracy values were found unsatisfactory results. Artificial intelligence, specifically deep learning, has the potential to give improved accuracy. This research employed deep learning to predict more accurate values of shrink–swell indices, which explored two scenarios; Scenario 1 used the features liquid limit, plastic limit, plasticity index, and linear shrinkage, whilst Scenario 2 added the input feature, fines percentage passing through a 0.075-mm sieve (%fines). Findings indicated that the implementation of deep learning neural networks resulted in increased model measurement accuracy in Scenarios 1 and 2. The values of accuracy measured in this study were suggestively higher and have wider variance than most previous studies. Global sensitivity analyses were also conducted to investigate the influence of each input feature. These sensitivity analyses resulted in a range of predicted values within the variance of data in Scenario 2, with the %fines having the highest contribution to the variance of the shrink–swell index and a relevant interaction between linear shrinkage and %fines. The proposed model Scenario 2 was around 10–65% more accurate than the preceding models considered in this study, which can then be used to expeditiously estimate more accurate values of shrink–swell indices.

Synthesising the spatial and temporal dynamics of the human body skeleton remains a challenging task, not only in terms of the quality of the generated shapes, but also of their diversity, particularly to synthesise realistic body movements of a specific action (action conditioning). In this paper, we propose Kinetic-GAN, a novel architecture that leverages the benefits of Generative Adversarial Networks and Graph Convolutional Networks to synthesise the kinetics of the human body. The proposed adversarial architecture can condition up to 120 different actions over local and global body movements while improving sample quality and diversity through latent space disentanglement and stochastic variations. Our experiments were carried out in three well-known datasets, where Kinetic-GAN notably surpasses the state-of-the-art methods in terms of distribution quality metrics while having the ability to synthesise more than one order of magnitude regarding the number of different actions. Our code and models are publicly available at https://github.com/DegardinBruno/Kinetic-GAN.

Over the years, unmanned aerial vehicles (UAVs) have been regarded as a potential solution to surveil public spaces, providing a cheap way for data collection, while covering large and difficult-to-reach areas. This kind of solutions can be particularly useful to detect, track and identify subjects of interest in crowds, for security/safety purposes. In this context, various datasets are publicly available, yet most of them are only suitable for evaluating detection, tracking and short-term re-identification techniques. This paper announces the free availability of the P-DESTRE dataset, the first of its kind to provide video/UAV-based data for pedestrian long-term re-identification research, with ID annotations consistent across data collected in different days. As a secondary contribution, we provide the results attained by the state-of-the-art pedestrian detection, tracking, short/long term re-identification techniques in well-known surveillance datasets, used as baselines for the corresponding effectiveness observed in the P-DESTRE data. This comparison highlights the discriminating characteristics of P-DESTRE with respect to similar sets. Finally, we identify the most problematic data degradation factors and co-variates for UAV-based automated data analysis, which should be considered in subsequent technologic/conceptual advances in this field. The dataset and the full specification of the empirical evaluation carried out are freely available at http://p-destre.di.ubi.pt/.

This article describes one objective function for learning semantically coherent feature embeddings in multi-output classification problems, i.e., when the response variables have dimension higher than one. Such coherent embeddings can be used simultaneously for different tasks, such as identity retrieval and soft biometrics labelling. We propose a generalization of the triplet loss that: 1) defines a metric that considers the number of agreeing labels between pairs of elements; 2) introduces the concept of similar classes, according to the values provided by the metric; and 3) disregards the notion of anchor, sampling four arbitrary elements at each time, from where two pairs are defined. The distances between elements in each pair are imposed according to their semantic similarity (i.e., the number of agreeing labels). Likewise the triplet loss, our proposal also privileges small distances between positive pairs. However, the key novelty is to additionally enforce that the distance between elements of any other pair corresponds inversely to their semantic similarity. The proposed loss yields embeddings with a strong correspondence between the classes centroids and their semantic descriptions. In practice, it is a natural choice to jointly infer coarse (soft biometrics) + fine (ID) labels, using simple rules such as k-neighbours. Also, in opposition to its triplet counterpart, the proposed loss appears to be agnostic with regard to demanding criteria for mining learning instances (such as the semi-hard pairs). Our experiments were carried out in five different datasets (BIODI, LFW, IJB-A, Megaface and PETA) and validate our assumptions, showing results that are comparable to the state-of-the-art in both the identity retrieval and soft biometrics labelling tasks.

The hair appearance is among the most valuable soft biometric traits when performing human recognition at-a-distance. Even in degraded data, the hair's appearance is instinctively used by humans to distinguish between individuals. In this paper we propose a multi-task deep neural model capable of segmenting the hair region, while also inferring the hair color, shape and style, all from in-the-wild images. Our main contributions are two-fold: 1) the design of an all-in-one neural network, based on depthwise separable convolutions to extract the features; and 2) the use convolutional feature masking layer as an attention mechanism that enforces the analysis only within the `hair' regions. In a conceptual perspective, the strength of our model is that the segmentation mask is used by the other tasks to perceive - at feature-map level - only the regions relevant to the attribute characterization task. This paradigm allows the network to analyze features from nonrectangular areas of the input data, which is particularly important, considering the irregularity of hair regions. Our experiments showed that the proposed approach reaches a hair segmentation performance comparable to the state-of-the-art, having as main advantage the fact of performing multiple levels of analysis in a single-shot paradigm.