I am an FNRS Chargée de Recherche at the LISA laboratory of the Université libre de Bruxelles, where I work on view synthesis, plenoptic imaging, and virtual reality. I build methods that recover and render real-world scenes directly from images, and I care about making those methods fast, robust, and useful outside the lab.

A recurring theme in my research is depth image-based rendering (DIBR) — the idea that a handful of views plus their depth maps is often enough to synthesize new ones, without the dense multi-view rigs that other approaches require. I developed the Reference View Synthesizer (RVS), a real-time DIBR tool released as open source, which is now the official view-synthesis reference tool in the MPEG-I standardization of immersive video. I have also built the companion RLC and RPVC tools for plenoptic 2.0 cameras and contributed to MPEG-I standardization since 2017.

Two extensions matter most to me. The first is non-Lambertian content: glossy, transparent, and reflective objects break the assumptions of classical DIBR because their features do not move linearly with the camera. My work on non-Lambertian maps models that non-linearity directly in the light field, letting such scenes be rendered and synthesized without an explicit 3D reconstruction. The second is plenoptic cameras: their dense local angular sampling changes what is possible. I have developed calibration and structure-from-motion methods that work on the raw micro-images of these cameras, and a view synthesizer that navigates free viewpoints from a single plenoptic capture.

My current agenda is to bring these lines together into a single framework for capturing and rendering real scenes that is physically grounded, runs in real time, and stays deployable on the hardware immersive systems actually use.

The About page has more on my background, and Google Scholar carries the full publication list.