header

Research Areas


Our research activities revolve around computational representations of the visual world, at the intersection of computer graphics, computer vision, signal processing, and machine learning. We seek to understand how visual information can be represented, processed, and interpreted efficiently, drawing on the synergies between physics-based modeling and modern data-driven approaches. A central focus of our research is the interaction of light with the world, which connects physical reality to visual observations and forms the foundation for both image synthesis and scene reconstruction.

We investigate novel representations of visual information, ranging from images and geometric structures to the appearance of complex three-dimensional scenes. Of particular interest are continuous representations, such as neural fields and Gaussian-based models, and their connections to classical signal processing. We explore how these representations can capture visual information across different scales, support complex light transport effects, and enable efficient rendering and analysis. Our work draws on mathematical modeling, learning-based techniques, and algorithmic optimization to develop representations that are expressive, scalable, and computationally efficient.

Another important direction of our research is the integration of prior knowledge through generative AI. Visual observations are often incomplete or ambiguous, making it difficult to infer the underlying properties of the world. We investigate how learned generative models can help resolve these ambiguities, complement physical models, and enable new ways of synthesizing and editing visual content. Across these research directions, we place particular emphasis on computational efficiency and interactivity, ultimately seeking to advance how we understand, synthesize, and explore the visual world.

For another perspective on our research, the word cloud below highlights keywords extracted from our publications.

VisRep research keyword cloud

Datenschutzerklärung/Privacy Policy Home Visual Computing institute RWTH Aachen University