We are specifically interested in hosting postdocs to conduct cutting-edge research on computational light field displays, computational microscopy, and optical neural networks.  Find below more detailed descriptions of the three research themes.

Computational light field displays

Light Field (LF) displays promise high-fidelity 3D scene recreation but remain constrained by physical and computational limitations. An MSCA PF project would address design challenges of LF displays by developing new diffractive optical elements that enable wide field of view while embedding optical computing and wavefront coding directly into the hardware. The project would advance the theory and simulation of light diffraction and complement the optical design with hybrid optical–neuromorphic computing architectures that replace energy‑intensive GPU pipelines with compact, hardware‑aware, brain‑inspired processing. The expected outcomes include a laboratory prototype, validated in near‑eye XR scenarios with respect to field of view, spatial‑angular resolution, energy efficiency, and user immersiveness.

For this project we are looking for ambitious postdoctoral researchers with a strong background in computational imaging, diffractive optics, or light-field formalism. Experience in neuromorphic computing and deep learning optimization is highly desirable. Candidates who have a track record in building optical prototypes or working with optical systems will be given particular consideration.

Computational microscopy

Optical (light) microscopy enables the acquisition of visual information about samples across spatial, spectral, and temporal dimensions in a non-invasive manner. Computational microscopy approaches, in particular, merge the design of hardware (light sources, optics, sensors) with computational algorithms for image reconstruction, enhancement, and analysis. This integration expands the information content beyond that of traditional microscopy and enables new functionalities such as snapshot 3D (volumetric) imaging and virtual staining of label-free tissue.

Future advancements in the field requires multidisciplinary research integrating accurate models of light–matter interaction with novel optical elements and advanced image processing algorithms. Crucially, these components must be jointly developed and tightly coupled to the specific problem or use case at hand.

We look forward to hosting ambitious postdocs to conduct research on developing novel computational optical microscopy methods. Particular topics of interest are light field (volumetric) fluorescence microscopy and virtual staining of label-free tissues. We appreciate a strong background in imaging, machine learning, and optics.

Optical neural networks

Optics offers a compelling alternative to digital electronics for neuromorphic systems, with Optical Neural Networks (ONNs) promising information processing that is orders of magnitude faster and more energy-efficient than conventional electronic approaches. However, large-scale deployment of ONNs is hindered by a mismatch between modern machine learning models and physical hardware constraints. In particular, the intrinsic linearity of optical systems limits expressive power compared to the nonlinear, nonlocal operations central to contemporary ML, while models trained in simulation often suffer performance degradation when transferred to real hardware due to the reality gap and limited in‑situ training capabilities.

We seek ambitious postdoctoral researchers to develop efficient ONNs addressing two key challenges: (i) enabling nonlinear information processing through minimal use of nonlinear optical elements via intelligent information encoding, and (ii) designing novel in‑situ and hardware-aware training algorithms for general optical processing and task-specific computer vision applications.

We value candidates with a strong background in machine learning, particularly in physics-informed neural networks and hardware-aware algorithm design. A demonstrated interest in experimental optics is considered a significant plus.

The research will be carried out within the 3D Media Group utilising the world-class infrastructure facilities of CIVIT. We offer a cross-disciplinary research environment, aligned with the university’s strong commitment to the field of imaging.