This study provides of a comparative analysis of emotion estimation from facial expressions under partial occlusion caused by face masks and extended reality (XR) headsets. Unlike previous studies that have independently explored these two scenarios, this research compares and analyzes the statistical differences between them. In order to achieve this, the RAF-DB dataset has been used as a non-occluded baseline to construct two new datasets: i) a dataset formed by faces partially occluded by face masks, and ii) a dataset formed by faces partially occluded by XR headsets. To evaluate the impact of occlusion in emotion estimation, three deep learning models have been fine-tuned using transfer learning, and results from a random classifier have been used as a baseline. Seven different metrics were obtained per dataset, and a 2-way ANOVA test was performed on each metric. As expected, significant statistical differences are observed between the non-occluded faces (acc. 0.8780) and the faces partially occluded by face masks (acc. 0.7520) and XR headsets (acc. 0.7400) on all metrics. Notably, the comparison between the two partially occluded datasets revealed significant statistical differences in the metrics f1-score (macro), precision (macro) and recall (macro), which we attribute to different types of occlusion affecting different parts of the face that are key to some emotions. This research contributes to advancing emotion recognition systems by highlighting their robustness and effectiveness even in partial occlusion settings, and showing a full comparative analysis between two common types of occlusion.
Markerless human-motion assessment has become increasingly relevant in domains such as psychomotor learning, rehabilitation, sports training, and collaborative educational environments. While recent pose-estimation frameworks provide reliable skeletal reconstruction, translating these data into interpretable posture assessment remains challenging, particularly under varying participant-to-sensor distances and in multi-participant settings. This paper presents the technological evolution from KUMITRON, a system originally developed for synchronized multi-participant monitoring and interaction analysis in psychomotor activities, to COLLVIT, a prototype implementing configurable depth-aware joint-angle posture assessment. This evolution encompasses RGB-D integration, configurable angular templates, depth-informed adaptive tolerance mechanisms, temporal stabilization strategies, interpretable posture assessment, and multi-participant processing capabilities. The paper distinguishes between the architectures defined at the patent level, their implementation through the KUMITRON and COLLVIT prototypes, and capabilities that remain targets for future implementation and validation. Rather than reporting a controlled experimental evaluation, the contribution of this work lies in documenting the technological evolution from KUMITRON to COLLVIT, the rationale behind the patent-based design decisions, the implementation of the current prototype, and directions for its future development and validation. The resulting framework provides a flexible foundation for future research on markerless kinematic posture assessment in educational, rehabilitation, and other psychomotor application domains.