Florent Lafarge

Inria
Titane research group
2004 routes des Lucioles
06902 Sophia Antipolis, France


email: florent(dot)lafarge(at)inria(dot)fr
ph:  +33 (0)4 92 38 77 31



I am a researcher at Inria in the Titane research group. I do research in Computer Vision, Geometry Processing and Remote Sensing. I work on the analysis and geometric modeling of 3D environments from physical measurements, typically multi-view imagery and laser scanning. My favourite topics include surface reconstruction and approximation, city modeling, piecewise-planar geometry and spatial point processes.

Research activities
Teaching
Publications
Codes, software & datasets


News
August: Invited keynote at ECCV2020 Holistic 3D on Data structures for piecewise-planar geometry [video (30min)]
August: Kinetic Shape Reconstruction presented at SIGGRAPH 2020 [video (15min)] [spotlight (30sec)]
June: Approximating shapes in images with low-complexity polygons presented at CVPR 2020 [oral presentation (5min)]
June: Connect-and-Slice: an hybrid approach for reconstructing 3D objects presented at CVPR 2020 [spotlight (1min)]


Selected publications

- 2020 -

Kinetic Shape Reconstruction
Bauchet J.P. and Lafarge F.
ACM Trans. on Graphics
, Vol. 39(5), 2020
[paper] [supplementary] [bibtex]


Approximating shapes in images with low-complexity polygons
Li M., Lafarge F. and Marlet R.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2020
[paper] [bibtex]


Connect-and-Slice: an hybrid approach for reconstructing 3D objects
Fang H. and Lafarge F.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2020
[paper] [suppl] [bibtex]


Extracting Geometric Structures in Images with Delaunay Point Processes
Favreau J.D., Lafarge F., Bousseau A. and Auvolat A.
IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI), Vol. 42(4), 2020
[paper] [bibtex]



- 2019 -

Low-power neural networks for semantic segmentation of satellite images
Bahl G., Daniel L., Moretti M. and Lafarge F.
Proc. of the ICCV Workshop on Low-Power Computer Vision (ICCVW)
, 2019
[paper] [bibtex]


City Reconstruction from Airborne Lidar: A Computational Geometry Approach
Bauchet J.P. and Lafarge F.
Proc. of 3D GeoInfo
, 2019
[paper] [bibtex]


Pyramid Scene Parsing Network in 3D: improving semantic segmentation of point clouds with multi-scale contextual information
Fang H. and Lafarge F.
ISPRS Journal of Photogrammetry and Remote Sensing (JPRS)
, Vol. 154, 2019
[paper] [bibtex]


A learning approach to evaluate the quality of 3D city models
Ennafii O., Le Bris A., Lafarge F. and Mallet C.
Photogrammetric engineering and remote sensing
, Vol. 85(12), 2019
[paper] [bibtex]



- 2018 -

KIPPI: KInetic Polygonal Partitioning of Images
Bauchet J.P. and Lafarge F.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2018
[paper] [bibtex]


Planar Shape Detection at Structural Scales
Fang H., Lafarge F. and Desbrun M.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2018
[paper] [bibtex]



- 2017 -

Photo2ClipArt: Image Abstraction and Vectorization Using Layered Linear Gradients
Favreau J.D., Lafarge F. and Bousseau A.
Proc. of SIGGRAPH ASIA conference
, 2017
[paper] [video] [bibtex]


Semantic segmentation of 3D textured meshes for urban scene analysis
Rouhani M., Lafarge F. and Alliez P.
ISPRS Journal of Photogrammetry and Remote Sensing (JPRS)
, Vol. 123, 2017
[paper] [bibtex]


Forest point processes for the automatic extraction of networks in raster data
Schmidt A., Lafarge F., Brenner C., Rottensteiner F. and Heipke C.
ISPRS Journal of Photogrammetry and Remote Sensing (JPRS)
, Vol. 125, 2017
[paper] [bibtex]



- 2016 -

Towards large-scale city reconstruction from satellites
Duan L. and Lafarge F.
Proc. of the European Conference on Computer Vision (ECCV)
, 2016
[paper] [bibtex] [supp.]


Fidelity vs. Simplicity: a Global Approach to Line Drawing Vectorization
Favreau J.D., Lafarge F. and Bousseau A.
Proc. of SIGGRAPH conference
, 2016
[paper] [bibtex] [webpage]


Object Classification via Planar Abstraction
Oesau S., Lafarge F. and Alliez P.
Proc. of the ISPRS Congress
, 2016
[paper] [bibtex]


Planar Shape Detection and Regularization in Tandem
Oesau S., Lafarge F. and Alliez P.
Computer Graphics Forum
, Vol. 35(1), 2016
[paper] [bibtex]



- 2015 -

Image Partitioning into Convex Polygons
Duan L. and Lafarge F.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2015
[paper] [bibtex]


Line Drawing Interpretation in a Multi-View Context
Favreau J.D., Lafarge F. and Bousseau A.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2015
[paper] [bibtex]


LOD Generation for Urban Scenes
Verdie Y., Lafarge F. and Alliez P.
ACM Trans. on Graphics
, Vol. 34(3), 2015
[paper] [bibtex]


Structure-Aware Mesh Decimation
Salinas D., Lafarge F. and Alliez P.
Computer Graphics Forum
, Vol. 34(6), 2015
[paper] [bibtex]



- 2014 -

Indoor Scene Reconstruction using Feature Sensitive Primitive Extraction and Graph-cut
Oesau S., Lafarge F. and Alliez P.
ISPRS Journal of Photogrammetry and Remote Sensing (JPRS)
, Vol. 90, 2014 (best paper award)
[paper] [bibtex]


Zometool Shape Approximation
Zimmer H., Lafarge F., Alliez P. and Kobbelt L.
Graphical Models
, Vol. 76(5), 2014
[paper] [bibtex] [video]


Detecting parametric objects in large scenes by Monte Carlo sampling
Verdie Y. and Lafarge F.
International Journal of Computer Vision (IJCV)
, Vol. 106(1), 2014
[paper] [bibtex]



- 2013 -

Recovering line-networks in images by junction-point processes
Chai D., Forstner W. and Lafarge F.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2013
[paper] [bibtex]


Surface reconstruction through point set structuring
Lafarge F. and Alliez P.
Proc. of Eurographics
, 2013
[paper] [bibtex]


A hybrid multi-view stereo algorithm for modeling urban scenes
Lafarge F., Keriven R.,Bredif M. and Vu H.
IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI), Vol. 35(1), 2013
[paper] [bibtex]



- 2012 -

Efficient Monte Carlo sampler for detecting parametric objects in large scenes
Verdie Y. and Lafarge F.
Proc. of the European Conference on Computer Vision (ECCV),
, 2012
[paper] [bibtex]


Creating large-scale city models from 3D-point clouds: a robust approach with hybrid representation
Lafarge F. and Mallet C.
International Journal of Computer Vision (IJCV),
Vol. 99(1), 2012
[paper] [bibtex]



- 2011 -

Stochastic geometry for image analysis
Chatelain F., Descombes X., Lafarge F., Lantuejoul C., Mallet C., Minlos M., Schmitt M., Sigelle M., Stoica R. and Zhizhina E.
Wiley/Iste,
Ed. X. Descombes, 2011
[html] [bibtex]


Building large urban environments from unstructured point data
Lafarge F. and Mallet C.
Proc. of the IEEE International Conference on Computer Vision (ICCV)
, 2011
[paper] [bibtex]



- 2010 -

Geometric Feature Extraction by a Multi-Marked Point Process
Lafarge F., Gimel'farb G. and Descombes X.
IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI),
Vol. 32(9), 2010
[paper] [bibtex]

A Marked Point Process for Modeling Lidar Waveforms
Mallet C., Lafarge F.,Roux M., Soergel U., Bretar F. and Heipke C.
IEEE Trans. on Image Processing (IP),
Vol. 19(12), 2010
[paper] [bibtex]

Hybrid Multi-view Reconstruction by Jump-Diffusion
Lafarge F., Keriven R.,Bredif M. and Vu H.
Proc. of the IEEE conference on Computer Vision and Pattern Recognition (CVPR)
, 2010
[paper] [bibtex] [presentation]

Insertion of 3D-Primitives in Mesh-Based Representations: Towards Compact Models Preserving the Details
Lafarge F., Keriven R. and Bredif M.
IEEE Trans. on Image Processing (IP),
Vol. 19(7), 2010
[paper] [bibtex]

Structural approach for building reconstruction from a single DSM
Lafarge F., Descombes X., Zerubia J. and Pierrot-Deseilligny M.
IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI),
Vol. 32(1), 2010
[paper] [bibtex]