Synthetic Rabbit Plenoptic Test Sequence

The test sequence "Synthetic Rabbit" is provided by Sarah Fachada, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, members of the LISA department, EPB (Ecole Polytechnique de Bruxelles), ULB (Universite Libre de Bruxelles), Belgium.

License: 

CC BY-NC-SA

ONLY Available for Academic Usage 

Terms of Use: 

Anykind of publication or report using this sequence should refer to the following references.

[1] Sarah Fachada, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, "Synthetic Rabbit Plenoptic Test Sequence", 2024.

@misc{fachada_syntheticrabbit_2024,    title = {{Synthetic} {Rabbit} {Plenoptic} {Test} {Sequence}},    author = {Fachada, Sarah and Bonatto, Daniele and Lafruit, Gauthier and Teratani, Mehrdad },    month = mar,    year = {2024},    doi = {10.5281/zenodo.10894522}}

Production:

Laboratory of Image Synthesis and Analysis, LISA department, Ecole Polytechnique de Bruxelles, Universite Libre de Bruxelles, Belgium.

Content:

This dataset contains a test scene acquired with synthetic plenoptic cameras and a line of 51 pinhole images. 

The dataset structure is the following:

"cameras.json" which is a json file with the parameters of the plenoptic cameras and of the pinhole cameras located at the main lens centers;

"cameras_nav.json" which is a json file with the parameters of a line of 51 pinhole cameras;

three folders, "Keplerian", "R5", "R8", each corresponding to a different plenoptic camera models, inspired respectively from Tsinghua University’s camera [1], Raytrix R5 and R8 models [2], each containing:

a "color" folder with images in plenoptic format. The images are in 4 versions, without degradation, with blur due to multi-focus, with luminance decay and with blur and luminance decay;

a "depth" folder with depth images associated to the color plenoptic image. The depth maps are in millimeters, from the main lens. We provide the ground truth depth maps, and depth maps obtained from Plenoptic Toolbox [3] and LLMV [4] disparity estimations;

a "pinhole" folder with the pinhole image placed at center of the main lens and its associated depth map;

a "navigation" folder with 51 color pinhole images placed every two millimeters.

References and links:

[1] X. Sun, X. Jin, T. Zhong, et P. Wang, "[MPEG-I Visual] A New Self-designed Focused Plenoptic Camera and the Captured Video Sequence “Boys” [m46259]", Marrakesh, Morocco, janv. 2019. doi: ISO/IEC JTC1/SC29/WG11 MPEG2019/M46259.

[2] Raytrix, https://raytrix.de/

[3] L. Palmieri, R. Koch, et R. O. H. Veld, "The Plenoptic 2.0 Toolbox: Benchmarking of Depth Estimation Methods for MLA-Based Focused Plenoptic Cameras", in 2018 25th IEEE International Conference on Image Processing (ICIP), IEEE, 2018. doi: 10.1109/ICIP.2018.8451073.

[4] D. Bonatto et al., "Multiview from Micro-Lens Image of Multi-Focused Plenoptic Camera", in 2021 International Conference on 3D Immersion, Brussels, Belgium, déc. 2021, p. 8. doi: 10.1109/IC3D53758.2021.9687243.

Acknoledgements:

Sarah Fachada is a postdoctoral researcher of the FNRSThis work was supported in part by the FER 2021 project (1060H000066-FAISAN), Belgium; in part by the Emile DEFAY 2021 project (4R00H000236), Belgium; and in part by the FER 2023 project (1060H000075), Belgium.