UnicornRandomCam

UnicornRandomCam:

This plenoptic sequence dataset "UnicornRandomCam" is provided by Eline Soetens, Gauthier Lafruit, Mehrdad Teratani members of the LISA department, EPB (Ecole Polytechnique de Bruxelles), ULB (Université Libre de Bruxelles), Belgium and Jonghoon Yim, Byeungwoo Jeon members of the Department of Electrical and Computer Engineering, SKKU (Sungkyunkwan University), Korea.

License:

CC BY-NC-SA

Terms of Use:

Any kind of publication or report using this sequence should refer to the following reference :

Eline Soetens, Jonghoon Yim, Byeungwoo Jeon, Gauthier Lafruit, Mehrdad Teratani, "UnicornRandomCam", 2023.

@misc{soetens_UnicornRandomCam_2023, title = {UnicornLinearCam}}, author = {Soetens, Eline and Yim, Jonghoon and Jeon, Byeungwoo and Lafruit, Gauthier and Teratani, Mehrdad }, month = sep, year = {2023}, doi = {10.5281/zenodo.8398535} }

Content:

We provide 30 lenslet images of a scene captured from different points of view. The scene was filmed with the plenoptic camera moved by hand then 30 frames of the sequence were selected. The images were captured using a Raytrix R8 plenoptic 2.0 camera [1] equipped with a 35mm lens. Each image has a 3840 x 2160 resolution. The scene contains five cubes of various sizes placed at different depths on a brown table. The sides of the cubes display a uniform color or simple pattern. A transparent sphere and a unicorn statue are placed among the cubes. The unicorn statue is highly reflective and presents a lot of specularities, thus both objects have a non-Lambertian texture.

Please find a detailed description of the content of each file below:

• lenslet_views: Contains the 30 lenslet images in PNG format.
• R8-Calibration_UnicornRandomCam.xml: Contains the calibration parameter of the plenoptic camera given by the Raytrix SDK [2].
• SubapertureConfig.cfg: The configuration file used to generate the 5x5 subaperture images using RLC 1.5 [3,4,5].
• center_views: Contains the center view of the subaperture images generated from each lenslet image.
• CamerasParameters.json: Extrinsic parameters for the 5x5x30 subaperture views generated from all lenslet images using RLC 1.5 and the config file provided. Parameters were obtained using the method described in [6,7,8].

Reference and links:

[1] https://raytrix.de/products/

[2] https://raytrix.de/downloads/

[3] Shu Fujita, Sho Mikawa, Mehrdad Teratani, Keita Takahashi, and Toshiaki Fujii, “Extracting Multi-View Images from Multi-Focused Plenoptic Camera,” in Proc. of International Workshop on Advanced Image Technology (IWAIT)/International Forum on Medical Imaging in Asia (IFMIA), in SPIE Digital Library, Singapore, January 2019.

[4] Mehrdad Teratani, Shu Fujita, Wenzhe Ouyang, Keita Takahashi, Toshiaki Fujii, "A 3-D Imaging System Using Multi-Focused Plenoptic Camera and Tensor Display," in Proc. of 2018 International Conference on 3D Immersion (IC3D2018) in IEEE Xplore Digital Library, Brussels, Belgium, December 2018.

[5] https://gitlab.com/mpeg-dense-light-field/rlc/-/tree/version1.5?ref_type=tags

[6] Sarah Fachada, Armand Losfeld, Takanori Senoh, Gauthier Lafruit, and Mehrdad Teratani, “A Calibration Method for Subaperture Views of Plenoptic 2.0 Camera Arrays,” in 2021 IEEE 23nd International Workshop on Multimedia Signal Processing, Tampere, Finland, Oct. 2021.

[7] Sarah Fachada, Armand Losfeld, Tankanori Senoh, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, "[LVC] [DLF] A new calibration tool for multi-plenoptic 2.0 cameras ", ISO/IEC JTC 1/SC 29/WG 4 m57837, Online, October 2021.

[8] https://gitlab.com/mpeg-dense-light-field/RPVC