Julie Artois successfully passed the public defense of her doctoral dissertation on March 11, 2026. The dissertation’s supervisors were Prof. Glenn Van Wallendael and Prof. Peter Lambert. The PhD thesis is titled “From Images to Immersion: Efficient Techniques for Light Field Rendering”.
Virtual reality and other immersive media have the potential to make users feel as if they are truly present inside a filmed scene. To enable this experience, a conventional video is not sufficient: the user must be able to look around freely and move through the scene. This PhD research investigates how dynamic real-world scenes can be captured using multiple synchronized cameras and subsequently reconstructed in digital form. This allows viewers to revisit and explore the original scene in a much more immersive way than is possible with traditional video.
The research focuses on efficient methods for converting the large amounts of captured data into interactive visualizations, for example, in virtual reality. The developed techniques can contribute to applications in entertainment, education, virtual presence, tourism, and cultural heritage, particularly in scenarios where visual realism and freedom of movement are of central importance.
Unfortunately, Julie’s public defense was not recorded. After the candidate’s presentation, the examination board deliberated in a private setting. Soon after, the chair of the examination committee announced the final decision. The jury unanimously awarded Julie Artois the academic degree of Doctor in Computer Science Engineering.
An electronic version of the thesis (196 pages, 127 MB) can be found here. The thesis includes high-level summaries in both Dutch and English
During her PhD research, Julie’s work resulted in the following publications:
Journal:
OpenDIBR: Open Real-Time Depth-Image-Based Renderer of Light Field Videos for VR (MTAP 2024)
MVS-Splatting: Fast Multi-View Stereo Depth Fusion for 3D Gaussian Splatting Initialization (IEEE Access 2025)
Conference:
This work received the Best Open Dataset and Software Paper Award at ACM MMSYS 2022.