Dimensionality Reduction for Real-Time Light-Field View Synthesis of Kernel-Based Models
Dimensionality Reduction for Real-Time Light-Field View Synthesis of Kernel-Based Models
Martijn is working as a PhD student at IDLAB-MEDIA. He obtained his degree in MSc in Computer Science in 2018, at the University of Ghent. The topic of his doctoral research is about leveraging mixture models to model and compress immersive light field scenes with high focus on real-time aspects and functionality such as streamability, 6-degrees-of-freedom interactivity, compression, and lightweight decoding. Martijn has a strong interest in high-quality and high-performance code. Other topics of interest are linear algebra, neural networks, image processing, GPU programming.
Favourite software technologies: C++, SDL2, Dear ImGui, Halide, bgfx. Reasonably likes Python and TensorFlow.
Dimensionality Reduction for Real-Time Light-Field View Synthesis of Kernel-Based Models
OpenDIBR is an openly-available wide-base ligth field renderer geared towards real-time display of multi-camera images/videos, even in Virtual Reality.
We're happy and honored to announce that our immersive plenoptic dataset paper "SILVR - A Synthetic Immersive Large-Volume Plenoptic Dataset" has been rewarded with the ACM MMSys 2022 Best Open Dataset and Software Paper.
SILVR, a dataset of light field images for six-degrees-of-freedom navigation in large fully-immersive volumes. The SILVR dataset is short for Synthetic Immersive Large-Volume Ray dataset.
Our paper "Keyframe Insertion for Random Access and Packet-Loss Repair in H.264/AVC, H.265/HEVC, and H.266/VVC" was accepted at the Data Compression Conference (DCC) 2022, and our paper "A study on keyframe injection in three generations of video coding standards for fast channel switching and packet-loss repair" was accepted at Multimedia Tools and Applications.
Telenet's .comdom app - safe sexting in a world of AI?
Two papers accepted for PCS 2018! Modeling and Real-Time Rendering of Light Field Video