1. Efficient material selection for training occluded mmWave radar-based gesture recognisersNuwan Attygalle, Luis A. Leiva, Matjaž Kljun, Klen Čopič Pucihar, 2026, izvirni znanstveni članek Opis: Radar-based gesture recognition has emerged as a promising approach for unobtrusive interaction. Unlike camera-based systems, radar sensors can detect gestures through opaque materials, enabling seamless embedding into various everyday objects. However, it remains unclear how to train models efficiently for robust gesture recognition through diverse materials. To investigate this, we collected a dataset of 17,520 gesture recordings performed through 73 everyday materials. By comparing several material-sampling and data-augmentation strategies, we found that a small carefully selected representative subset of training materials was sufficient to match performance of a classifier trained on the full material dataset. Our results showed that the models trained on 14 quota-sampled materials achieved accuracies of 95.8% and 91.2%, comparable to training on all 73 materials (96.8% and 91.6%) and significantly better than training without material data (66.8% and 65.8%). Among the evaluated sampling approaches, Quota sampling also provided the best overall trade-off between performance and practicality. In contrast, classifiers trained on augmented data performed worse than those trained on actual material-specific data. Taken together, these findings indicate that, for the tested sensor, gesture set, and material collection, carefully selected real-material data offer a practical route to reducing material-specific data collection in radar-based gesture recognition while preserving generalisation. Code, models, and data are available in the public repository, with additional details provided in the supplementary materials: https://gitlab.com/hicuplab/seeing-through.
Ključne besede: deep learning, training optimization for neural networks, millimetre-wave radar, gesture recognition, material selection, sampling, synthetic noise
Objavljeno v RUP: 19.06.2026; Ogledov: 229; Prenosov: 9
 Celotno besedilo (15,56 MB)
Gradivo ima več datotek! Več... |
2. Inferring a Mobile User’s Valence and Arousal through On-Screen Text AnalysisEdita Džubur, Veljko Pejović, 2025, samostojni znanstveni sestavek ali poglavje v monografski publikaciji Opis: Understanding a user’s emotional state is critical for building adaptive and intelligent mobile applications. In this paper we investigate the feasibility of inferring valence and arousal from the text displayed on smartphone screens. We developed AV-Sense, a mobile application that combines the Experience Sampling Method, a technique that prompts users to report their feelings in the moment, with passive screentext logging. In a two-week study with 12 participants, we collected 787 ESM responses and over 650,000 screentext entries. Data analysis revealed meaningful temporal and individual patterns in reported affect. We then explored the use of large language models to predict valence and arousal from screentext, but results indicated limited predictive power in this setting. Our findings highlight both the potential and current challenges of screentext-based affect inference, laying the groundwork for future research on emotion-aware applications and naturalistic psychological studies.
Ključne besede: text analysis, experience sampling method, screentext sensing, valence, arousal, large language models
Objavljeno v RUP: 30.01.2026; Ogledov: 769; Prenosov: 3
Celotno besedilo (350,11 KB) |
3. |
4. The status of a rare phylogeographic lineage of the vulnerable European souslik Spermophilus citellus, endemic to central MacedoniaBoris Kryštufek, Peter Glasnović, Svetozar Petkovski, 2012, drugi znanstveni članki Opis: The conversion of grasslands for agriculture has triggered a serious decline of the European ground squirrel or souslik Spermophilus citellus, categorized as Vulnerable on the IUCN Red List since 1996. The Jakupica phylogeographic lineage of central Macedonia is the smallest of the three major evolutionary lines of the European souslik. This lineage is an important reservoir of within-species diversity and should be regarded as an independent unit for conservation management purposes. It is endemic to Mount Jakupica, where it lives in mountain pastures at 1,500-2,250m altitude. The total area occupied by sousliks (884 ha) is fragmented and 94% of individuals occur in four colonies. Densities (0.8-5.5 adults ha-1) are lower than those reported elsewhere for the species, with the total population probably ,2,000 adults. One large colony, reportedly of c. 1,000 sousliks, was decimated in 2007 by a catastrophic fire and had still not recovered bz 2010. A steady decline in livestock grazing, together with the predicted advance of the tree line as a consequence of climate change, will probably reduce the optimal habitat for the souslik and negatively affect population fitness. Monitoring needs to be implemented, at least for the largest colonies, to provide early warning of any declines
Ključne besede: Spermophilus citellus, souslik, European souslik, density, endemic, distance sampling, fragmentation, Macedonia
Objavljeno v RUP: 15.10.2013; Ogledov: 6325; Prenosov: 84
 Povezava na celotno besedilo |
5. Genetic structure and evidence for recent population decline in Eurasian otter populations in the Czech and Slovak RepublicsPetra Hájková, Cino Pertoldi, Barbora Zemanová, K. Roche, B. Hájek, Josef Bryja, Jan Zima, 2007, izvirni znanstveni članek Ključne besede: otter, Lutra lutra, population structure, microsatellites, effective population size, genetic sampling, conservation management, Czech Republic, Slovak Republic
Objavljeno v RUP: 15.10.2013; Ogledov: 5905; Prenosov: 65
Povezava na celotno besedilo |
