| Session: | Content preparation and compression for VR |
| Location: | Lecture Room |
| Session Time: | Wednesday, June 27, 10:20 - 12:40 |
| Presentation Time: | Wednesday, June 27, 12:00 - 12:20 |
| Presentation: |
Special Session Lecture
|
| Paper Title: |
HARD REAL-TIME, PIXEL-PARALLEL RENDERING OF LIGHT FIELD VIDEOS USING STEERED MIXTURE-OF-EXPERTS |
| Authors: |
Ignace Saenen; Ghent University - imec, IDLab, Belgium | | |
| | Ruben Verhack; Ghent University - imec, IDLab / TU Berlin - Communication Systems Lab, Belgium | | |
| | Vasileios Avramelos; Ghent University - imec, IDLab, Belgium | | |
| | Glenn Van Wallendael; Ghent University - imec, IDLab, Belgium | | |
| | Peter Lambert; Ghent University - imec, IDLab, Belgium | | |
| Abstract: |
Steered Mixture-of-Experts (SMoE) is a novel framework for the approximation, coding, and description of image modalities such as light field images and video. The future goal is to arrive at a representation for Six Degrees-of-Freedom (6DoF) image data. Previous research has shown the feasibility of real-time pixel-parallel rendering of static light field images. Each pixel is independently reconstructed by kernels that lay in its vicinity. The number of kernels involved forms the bottleneck on the achievable framerate. The goal of this paper is twofold. Firstly, we introduce pixel-level rendering of light field video, as previous work only rendered static content. Secondly, we investigate rendering using a predefined number of most significant kernels. As such, we can deliver hard real-time constraints by trading off the reconstruction quality. |