Doctoral colloquium - Martin Halaj (23.3.2026)
Monday 23.3.2026 at 13:10, Lecture room I 9
Martin Halaj:
Beyond Static Geometry: Integrating Metrically Accurate Avatars into Dynamic Scenarios
Abstract:
The generation of realistic and metrically accurate 3D human models plays a pivotal role across a wide range of applications, including virtual and augmented reality, biomechanics, and healthcare. While recent advances have enabled the increasingly precise reconstruction of human body geometry, the transition from static models to dynamic human representation introduces new challenges. In this context, motion synthesis and synthetic data generation emerge as essential components. This research explores methods for generating plausible human motion and discusses the development of synthetic datasets designed for behavioral analysis, with a particular emphasis on scenarios involving violent and dangerous activities. Such datasets aim to overcome the inherent limitations of real-world data, including scarcity, annotation complexity, and ethical constraints.