Bernhard Kerbl

Bernhard Kerbl is a Post-doctoral Researcher at the Computer Graphics Group (E193-02) within the Faculty of Informatics at Vienna University of Technology (TU Wien). He actively contributes to research in real-time rendering, GPU optimization, and 3D visualization. His work focuses on point cloud rendering, Gaussian splatting, and low-level graphics API development (particularly Vulkan).

• Core Research Areas: Real-time rendering, GPU programming, point cloud processing, 3D Gaussian splatting, task-based parallelism
• Projects: IVILPC (2023–2026), ACD (2020–2028), EVOCATION (2018–2022)

Key Contributions: Developed real-time rendering techniques for massive datasets, pioneered GPU-based scene streaming architectures, and created novel methods for visual error prediction using machine learning. His work on Vulkan transition in academia received recognition, and he maintains active collaborations in GPU education.

Scientific Achievements:

• Best Paper Award at EGPGV 2024 for "Fast Rendering of Parametric Objects on Modern GPUs"
• Recipient of Vienna Science and Technology Fund (WWTF) support for graphics research

Technical Expertise: Specializes in CUDA, OpenGL/Vulkan APIs, and real-time systems. His work bridges academic research with practical implementations in virtual reality, augmented reality, and GPU education.