SIBR datasets available for download




README


===========================================================
Datasets for IBR algorithms using the SIBR library
===========================================================

We provide data preprocessed for each dataset in different packages. For instructions on downloading and using the software please see  https://sibr.gitlabpages.inria.fr/  and  https://gitlab.inria.fr/sibr/sibr_core .

* "Preproc. data for all algos" has data allowing to run all five algorithms [Chaurasia 13, Ortiz-Cayon 15, Hedman 16, Hedman 18] and [Buehler 01]
* "Unstructured Lumigraph" allow to run our implementation of per-pixel Unstructured Lumigraph [Buehler 01]
* "Deep Blending viewer only" allows to run the [Hedman 18] algorithm (viewer only)
* "Depth Synthesis Superpixel Warp and Selective Rendering" allows to run [Chaurasia 13] and [Ortiz-Cayon 15]
* "Colmap data" provides the entire colmap directory allowing to run Deep Blending and Inside Out preprocessing
* "Inside Out" allows to run the [Hedman 16] algorithm re-implemented as a baseline in [Hedman 18] (only MVS input). For this you must have downloaded either "Colmap" or "Deep Blending viewer only"
* "Deep Blending (preproc)" provides additional data needed to run the preprocessing of [Hedman 18]


===========================================================

References:

[Buehler 01] C. Buehler, M. Bosse, L. McMillan, S. Gortler, and M. Cohen. "Unstructured lumigraph rendering." In Proceedings SIGGRAPH 2001, pp. 425-432. ACM, 2001. https://www.ics.uci.edu/~gopi/ICS280Win02/UnstructuredLumigraph.pdf

[Chaurasia 13] Depth synthesis and local warps for plausible image-based navigation, (http://www-sop.inria.fr/reves/Basilic/2013/CDSD13/)

[Ortiz-Cayon 15] A Bayesian approach for selective image-based rendering using superpixels (http://www-sop.inria.fr/reves/Basilic/2015/ODD15/).

[Hedman et al. 16] Scalable Inside-Out Image-Based Rendering, (http://www-sop.inria.fr/reves/Basilic/2016/HRDB16)

[Hedman et al. 18] Deep Blending for Free-Viewpoint Image-Based Rendering, http://www-sop.inria.fr/reves/Basilic/2018/HPPFDB18)