[go: up one dir, main page]

AliceVision Association

The ALICEVISION association is a non-profit organization whose ambition is to democratize 3D digitization technologies from photographs.

  Make a Donation

Follow us

Mission

The association considers 3D digitization technologies essential for many applications for the common good, especially for culture and artistic creation.
The association's mission is to carry out the following scientific and educational actions:

Promoting the development of free computer vision software, respectful of open standards, that can be used or modified freely by all
Promote open and reproducible research allowing for transparent testing and verification of results
Promoting access to research knowledge and data
Promoting these technologies to industry, research, education and the general public

In particular, the association supports the development and promotion of open source software tools grouped in the AliceVision project.

Projects

The ALICEVISION association is developing 2 main open source projects: AliceVision framework and Meshroom software.

AliceVision Framework

AliceVision framework provides computer vision tools and software libraries to enable 3D reconstructions from photographs taken with all types of cameras, from professional cameras to smartphones. These photogrammetry tools are 100% free and accessible to all. The solution takes advantage of the GPU for some of the most expensive computational steps. The proposed solution is scalable in terms of the number of images - it is capable of processing datasets of more than 1,000 images. The framework relies on standard and open-source file formats (OpenEXR, Alembic, OBJ) to facilitate the interoperability of the software with other professional tools. Integrations are offe#EE6767 with content creation tools such as Blender®, Maya® (Autodesk), Nuke® / Mari® (Foundry) and Houdini® (SideFX).

Meshroom Software

Meshroom software is a complete photogrammetry solution built around AliceVision. It provides a default workflow for standard use cases, a nodal editor with a caching mechanism for the files created at each step, a post-processing of the created mesh and a possible retexturing after retopology. Meshroom allows users to calculate their 3D reconstruction locally or on a render farm. The nodal interface allows users to adapt the reconstruction pipeline to their production needs, finding a compromise between calculation time and quality of the result. Meshroom has been used since 2010 in the creation of digital environments for the visual effects industry (feature films and commercials), as well as since a few years in other industries such as cosmetics, sports manufacturing, augmented reality surgery, real estate…

The ALICEVISION project also includes a library for SIFT detection on GPU and an original markers detector called CCTag with CPU and GPU implementations. The association also contributes to the open source ecosystem it relies on, like OpenImageIO, Vcpkg, etc.

A Non-Profit Organization

The ALICEVISION association is governed by the French law of 1st July 1901. It is therefore a voluntary organization without any distribution of profits, neither for its leaders nor for its members. The ALICEVISION association, like any association, is a legal entity with its own legal capacity and responsibility, distinct from the persons managing it.
Voluntary work is the rule concerning the association : all donations and revenues are kept and used by the legal entity, without any distribution to the managers or members, despite their work and investment.

The Board of Directors of the ALICEVISION association is currently composed by:

    Benoit Maujean, President
    Alexandre Jenny, Secretary General
    Simone Gasparini, Treasurer
    Fabien Castan

Members

The ALICEVISION association has been founded by 18 people from European research and industry, who have been collaborating since 2010 around 3D computer vision for cultural and creative industries.

History

In 2010, IMAGINE research team (a joint research group between Ecole des Ponts ParisTech and Centre Scientifique et Technique du Bâtiment) and Mikros Image started a partnership around Pierre Moulon’s thesis, supervised by Renaud Marlet and Pascal Monasse on the academic side and Benoit Maujean on the industrial side. In 2013, they released an open source SfM pipeline, called openMVG (“Multiple View Geometry”), to provide the basis of a better solution for the creation of visual effects matte-paintings.

In 2009, CMP research team from CTU started the PhD thesis of Michal Jancosek supervised by Tomas Pajdla. They released windows binaries of their MVS pipeline, called CMPMVS, in 2012.

In 2009, Toulouse INP (Lilian Calvet, Vincent Charvillat, Pierre Gurdjos), INRIA (Simone Gasparini, Peter Sturm) and Duran Duboi (Fabien Castan, Eloi Du Bois) started a French ANR project to create a model based Camera Tracking solution based on natural features and a new marker design called CCTag.

In 2015, Simula (Lilian Calvet, Håvard Nygaard Espeland, Carsten Griwodz), Toulouse INP (Vincent Charvillat, Simone Gasparini, Pierre Gurdjos) and Mikros Image (Fabien Castan, Benoit Maujean, Jean Mélou, Cyril Pichard, Nicolas Rondaud) joined their efforts in the EU project POPART to create a Previz system.

In 2015, Michal Polic started a PhD at CTU on uncertainty estimation (supervised by Tomas Pajdla).

In 2017, CTU (Tomas Pajdla, Michal Polic) joined the partners group, with Simula (Carsten Griwodz, Konstantin Pogorelov), Mikros (Fabien Castan, Yann Lanthony, Gregoire De Lillo, Benoit Maujean), Toulouse INP (Vincent Charvillat, Clément Debize, Simone Gasparini, Pierre Gurdjos, Hatem Rashwan), in the EU project LADIO to create a central hub with access to all data generated on set.

In 2017, Jean Mélou started a PhD (CIFRE contract) with Toulouse INP (supervised by Jean Denis-Durou and Yvain Queau) and Mikros.

In 2020, INP and Mikros joined their effort to create the collaborative lab ALICIA-Vision, supported by French Agence Nationale de la Recherche.

In 2020, the ALICEVISION association is created.

In 2021, ALICEVISION has been awarded an Epic Games MegaGrant.

In 2021, Carsten Griwodz, Simone Gasparini, Lilian Calvet, Pierre Gurdjos, Fabien Castan, Benoit Maujean, Yann Lanthony and Gregoire De Lillo have been awarded to Best Open Dataset and Software Paper Award at MMSys 21 for the publication "AliceVision Meshroom: An Open-Source 3D Reconstruction Pipeline".

Milestones

  •    2015 : First open source version of Meshroom (SfM only)
  •    2016 : First open source version of CCTag with CPU and GPU implementations
  •    2016 : First open source version of PopSift
  •     2018 : AliceVision / Meshroom conference at FMX 2018 (Stuttgart)
  •    2018 : Meshroom first release as a complete 3D reconstruction pipeline (with both SfM and MVS)
  •     2018 : AliceVision Bird of a Feather at Siggraph 2018 (Vancouver)
  •    2019 : New versions of Meshroom (2019.1 and 2019.2) with a new user interface, improved mesh and texture.
  •     2019 : Meshroom Bird of a Feather at Siggraph 2019 (Los Angeles)
  •     2020 : AliceVision session at ASWF Open Source Days (during online Siggraph)
  •    2020 : New version of Meshroom (2020.1) with new HDR 360° Panorama pipeline
  •    2021 : New version of Meshroom (2021.1) with improved Panorama pipeline and new graphic editor UX
  •     2021 : Meshroom session at FMX 2021 Open Source track
  •     2021 : AliceVision session at Eurographics 2021 Volumetric Video Tutorial
  •    2023 : New version of Meshroom (2023.1) with new Color Management, improved 3D Reconstruction and Panorama pipelines

Contribute

ALICEVISION project relies on an open and friendly community of contributors always enthusiastic to share new ideas and start new collaborations.

If you are a Computer Vision Researcher, do not hesitate to reach us. We are always interested in discussing new approaches and comparing different solutions.

If you are a Developer and you want to contribute to the development of the software, do not hesitate to checkout the github and have a look at CONTRIBUTING.md for the first steps. There are a huge amount of possible improvements, regarding performance, user interface, documentation, unit tests, CI/CD etc.

If you are a 3D scanning or photogrammetry expert and you have some time to share, do not hesitate to join and contribute to the ALICEVISION online forum. The community will also appreciate your contribution to the online documentation.