Edge Rewrite
pun Source code View original
Jump to content

Draft:Florent Poux

From Wikipedia, the free encyclopedia

Florent Poux
Dr. Florent Poux, Head Professor of the 3D Geodata Academy, and Associate Researcher at the University of Liège.
EducationConservatoire national des arts et métiers (Civil Engineer)
University of Liège (PhD)
RWTH Aachen University (Postdoc)
Known forSmart Point Cloud infrastructure; 3D Data Science with Python
AwardsEuroSDR PhD Award (2019)
ISPRS Jack Dangermond Award (2017–2019)
Scientific career
FieldsGeomatics, photogrammetry, point cloud processing
WorkplacesUniversity of Liège
RWTH Aachen University
3D Geodata Academy
ThesisThe Smart Point Cloud: Structuring 3D Intelligent Point Data (2019)
Roland Billen
Websitelearngeodata.eu

Florent Poux is a researcher, educator and author in geomatics whose work concerns the automatic processing and semantic interpretation of three-dimensional point clouds captured by lidar and photogrammetry. He received a doctorate from the University of Liège in 2019, and a paper he wrote with his doctoral advisor Roland Billen on voxel-based point cloud segmentation received the Jack Dangermond Award, presented under the auspices of the International Society for Photogrammetry and Remote Sensing (ISPRS). He founded the 3D Geodata Academy, an online training company, and is the author of 3D Data Science with Python (O'Reilly Media, 2025).

Education

[edit]

Dr. Florent Poux began his career in land surveying.[1] He completed an engineering degree at the Conservatoire national des arts et métiers (CNAM), where his 2013 engineering dissertation, supervised by Jean-Michel Follin and hosted by the University of Liège, examined the optimisation and automation of 3D model production from laser scanning.[2]

He then prepared a doctorate in sciences at the Department of Geography of the University of Liège under the supervision of Roland Billen. His thesis, The Smart Point Cloud: Structuring 3D Intelligent Point Data, was defended on 5 June 2019.[3]

Career

[edit]

Dr. Florent Poux was a member of the geomatics research unit of the University of Liège from 2015, first while completing his doctorate, and later held the title of adjunct professor in 3D geodata there.[4][5][1] After his doctorate he carried out postdoctoral research in geometric computer vision and geometric deep learning at RWTH Aachen University, in the computer graphics group led by Leif Kobbelt.[1][6]

According to LIDAR Magazine, Dr. Florent Poux left academia in 2021 to start the 3D Geodata Academy.[1]

3D Geodata Academy

[edit]

The 3D Geodata Academy provides online courses on 3D data processing, including point cloud processing, photogrammetry and 3D deep learning, taught mainly with the Python programming language.[7] The company, 3D Geodata Academy SARL, was registered in Paris on 10 January 2025 and is listed as a training provider (organisme de formation) in the French business register.[8]

Research

[edit]

Dr. Poux's research addresses the automation of point cloud processing, in particular segmentation, classification and the connection of 3D data to formalised domain knowledge. In a 2016 paper with colleagues at Liège he set out a definition of the "smart point cloud" and the challenges remaining for its implementation.[9] His doctoral thesis developed this into a Smart Point Cloud infrastructure, a modular architecture that links point cloud properties with domain knowledge represented as graphs, so that relevant information can be extracted automatically and used by reasoning systems.[3]

Related work with Liège colleagues proposed a framework for the semantic modelling of indoor spaces and furniture from point clouds,[10] and compared unsupervised voxel-based segmentation using geometric and relationship features with deep learning methods.[11] At RWTH Aachen he co-authored work on a virtual reality system for cultural heritage and digital tourism,[12] and on an automatic region-growing system for segmenting large point clouds.[13] He also co-authored a review of 3D change detection methods based on point clouds.[14]

Awards

[edit]
  • 2019: EuroSDR PhD Award, from European Spatial Data Research (EuroSDR), for his doctoral thesis.[15]
  • Jack Dangermond Award for the 2017–2019 period, shared with Roland Billen, for their paper on voxel-based point cloud semantic segmentation in the ISPRS International Journal of Geo-Information. The award, named after Jack Dangermond, is presented every four years by the journal's publisher MDPI and by Esri with an ISPRS certificate. The same paper received the journal's Best Paper Award for 2019.[16][17]

Books

[edit]

Poux is the author of 3D Data Science with Python: Building Accurate Digital Environments with 3D Point Cloud Workflows, published by O'Reilly Media as an e-book on 9 April 2025 and in paperback in May 2025.[18] The book covers 3D data representations, 3D reconstruction, supervised and unsupervised machine learning for 3D data, and applications in computer vision, geospatial analysis and robotics, using Python libraries.[19] The documentation of the open-source Open3D library features the book.[20] He also publishes tutorials on 3D data processing in the online publication Towards Data Science.[21]

Selected publications

[edit]
  • Poux, Florent (2025). 3D Data Science with Python: Building Accurate Digital Environments with 3D Point Cloud Workflows. Sebastopol, CA: O'Reilly Media. ISBN 978-1-0981-6133-0.
  • Poux, Florent; Billen, Roland (2019). "Voxel-based 3D Point Cloud Semantic Segmentation: Unsupervised Geometric and Relationship Featuring vs Deep Learning Methods". ISPRS International Journal of Geo-Information. 8 (5) 213. doi:10.3390/ijgi8050213.
  • Poux, Florent; Neuville, Romain; Nys, Gilles-Antoine; Billen, Roland (2018). "3D Point Cloud Semantic Modelling: Integrated Framework for Indoor Spaces and Furniture". Remote Sensing. 10 (9) 1412. doi:10.3390/rs10091412.
  • Poux, F.; Mattes, C.; Selman, Z.; Kobbelt, L. (2022). "Automatic region-growing system for the segmentation of large point clouds". Automation in Construction. 138 104250. doi:10.1016/j.autcon.2022.104250.

References

[edit]
  1. 1 2 3 4 Austin Madson (13 October 2025). "Florent Poux". LIDAR Magazine (Podcast). No. 24. Retrieved 11 September 2026.
  2. ↑ Poux, Florent (2013). Vers de nouvelles perspectives lasergrammétriques : optimisation et automatisation de la chaîne de production de modèles 3D (Engineering thesis) (in French). Conservatoire national des arts et métiers. Retrieved 11 September 2026.
  3. 1 2 Poux, Florent (2019). The Smart Point Cloud: Structuring 3D Intelligent Point Data (Doctorat en Sciences thesis). Université de Liège. hdl:2268/235520. Retrieved 11 September 2026.
  4. ↑ "Florent Poux". ORCID. Retrieved 11 September 2026.
  5. ↑ "Florent Poux". GeoScITY, Université de Liège. Retrieved 11 September 2026.
  6. ↑ "Florent Poux". Computer Graphics and Multimedia, RWTH Aachen University. Retrieved 11 September 2026.
  7. ↑ "About". 3D Geodata Academy. Retrieved 11 September 2026.
  8. ↑ "3D GEODATA ACADEMY (SIREN 939 586 137)". Annuaire des Entreprises (in French). Government of France. Retrieved 11 September 2026.
  9. ↑ Poux, F.; Hallot, P.; Neuville, R.; Billen, R. (2016). "Smart Point Cloud: Definition and Remaining Challenges". ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences. IV-2/W1: 119–127. doi:10.5194/isprs-annals-IV-2-W1-119-2016.
  10. ↑ Poux, Florent; Neuville, Romain; Nys, Gilles-Antoine; Billen, Roland (2018). "3D Point Cloud Semantic Modelling: Integrated Framework for Indoor Spaces and Furniture". Remote Sensing. 10 (9) 1412. doi:10.3390/rs10091412.
  11. ↑ Poux, Florent; Billen, Roland (2019). "Voxel-based 3D Point Cloud Semantic Segmentation: Unsupervised Geometric and Relationship Featuring vs Deep Learning Methods". ISPRS International Journal of Geo-Information. 8 (5) 213. doi:10.3390/ijgi8050213.
  12. ↑ Poux, Florent; Valembois, Quentin; Mattes, Christian; Kobbelt, Leif; Billen, Roland (2020). "Initial User-Centered Design of a Virtual Reality Heritage System: Applications for Digital Tourism". Remote Sensing. 12 (16) 2583. doi:10.3390/rs12162583.
  13. ↑ Poux, F.; Mattes, C.; Selman, Z.; Kobbelt, L. (2022). "Automatic region-growing system for the segmentation of large point clouds". Automation in Construction. 138 104250. doi:10.1016/j.autcon.2022.104250.
  14. ↑ Kharroubi, Abderrazzaq; Poux, Florent; Ballouch, Zouhair; Hajji, Rafika; Billen, Roland (2022). "Three Dimensional Change Detection Using Point Clouds: A Review". Geomatics. 2 (4): 457–485. doi:10.3390/geomatics2040025.
  15. ↑ "EuroSDR Award winner 2019". EuroSDR. 8 October 2019. Retrieved 11 September 2026.
  16. ↑ "The Jack Dangermond Award". International Society for Photogrammetry and Remote Sensing. Retrieved 11 September 2026.
  17. ↑ "Florent Poux and Roland Billen, winners of the 2019 Jack Dangermond Award". Université de Liège. 13 May 2020. Retrieved 11 September 2026.
  18. ↑ "3D Data Science with Python (paperback)". Target. Retrieved 11 September 2026.
  19. ↑ Poux, Florent (2025). 3D Data Science with Python: Building Accurate Digital Environments with 3D Point Cloud Workflows. O'Reilly Media. ISBN 978-1-0981-6129-3.
  20. ↑ "3D Data Science with Python". Open3D documentation. Retrieved 11 September 2026.
  21. ↑ "Florent Poux". Towards Data Science. Retrieved 11 September 2026.
[edit]

Category:Living people Category:Year of birth missing (living people) Category:Conservatoire national des arts et métiers alumni Category:University of Liège alumni Category:Academic staff of the University of Liège Category:Photogrammetrists Category:Computer vision researchers

Retrieved from "https://en.wikipedia.org/w/index.php?title=Draft:Florent_Poux&oldid=1375224054"