Multimedia Information Processing Group

Plenoptic Cameras - Analysis and Simulation

The main aim of the project is to create and understand data from so-called light field or plenoptic cameras with a focus on the different types of plenoptic cameras: Plenoptic 1.0 (Lytro Cameras), Plenoptic 2.0 (Raytrix Cameras as shown below) and Plenoptic Microscopy (FiMic).

Raytrix Cameras

 

 

 

 

 

 

This research project includes several topics (Click on images to enlarge):

  •  
Depth estimation: One goal is the development of efficient and high-quality methods to estimate the depth map from plenoptic images as shown in the image below. More information regarding this aspect and source code with images are available at https://github.com/PlenopticToolbox/PlenopticToolbox2.0.
 
  plenoptic depth image
 
  •  
Dataset Aquisition: A new dataset was created capturing the same scene under the same conditions with both plenoptic 1.0 (Lytro Illum) and plenoptic 2.0 (Raytrix R29) cameras. More information and data available at: https://figshare.com/articles/The_Plenoptic_Dataset/6115487
 
  dataset
 
  •  
Simulation: We also aim to simulate the different plenoptic cameras in order to produce ground truth data for the evaluation of algorithms concerning e.g. the calibration or depth estimation. Additional information and Blender models are available at https://github.com/Arne-Petersen/Plenoptic-Simulation.
 
  Example
Image of a scene rendered via our Blender plenoptic camera model.
 

 

Project Partners and Funding

European Training Network on Full Parallax Imaging
supported by the Marie Skłodowska-Curie actions under the EU Research and Innovation programme Horizon 2020
Project website: www.full-parallax-imaging.eu

eu logo

Raytrix GmbH
Website: www.raytrix.de

raytrix logo