IMAGINE is a research group in
computer vision, machine learning and optimization of
the École des Ponts ParisTech
(a.k.a. ENPC).
It is affiliated to
the A3SI team of the LIGM computer science lab, in
Paris-Est Sup.
Open Positions
No open position at the moment (internship, PhD, ...).
For any application, please send:
- a cover letter explaining your interest and adequacy for a/the topic,
- your CV/resume,
- transcripts of your grades from last year, as well as this year when already available.
Recommendation letters are a plus, but are not compulsory at internship level.
Research
After having worked on programming languages, software engineering,
security and natural language processing, I have turned to
computer vision. I have joined the IMAGINE group in December 2009, and Valeo.ai in June 2019 (staying part-time in IMAGINE). I have been also a member of the Astra-vision group at Inria since 2022.
I am interested in the reconstruction of 3D models from images and range data, regarding both geometry and semantics, in particular with applications to building and city modeling:
- Camera registration. I have been working on external camera calibration issues (structure from motion), with a focus on accuracy and robustness, developing adaptive parameterless methods and global registration techniques, based on feature points as well as line segments. This also involves robust feature matching.
- Geometry processing. I have addressed speed and robustness issues for the treatment of point clouds, in particular regarding normal estimation for scenes with sharp features, and I have proposed various surface reconstruction methods, from simple watertight piecewise-planar polygonal meshes to finer and more accurate volumes, with or without learned priors.
- Scene understanding. I have been working on various methods to semantically segment 2D and 3D data, using grammar-based approaches (top-down or bottom-up, with handwritten grammars or learned priors) as well as less structured but more accurate learning-based approaches.
I was the coordinator (PI) of ANR project Semapolis (2013-2017) on the Semantic Visual Analysis and 3D Reconstruction of Urban Environments.
More recently, I have also been interested in computer vision for robotic applications in the context of civil engineering as well as safe driving and autonomous vehicles, using both 2D and 3D data.
Selected Publications in Computer Vision & Geometry Processing
See also publications on Google Scholar.
- ManiPose: Manifold-Constrained Multi-Hypothesis 3D Human Pose Estimation.
Cédric Rommel, Victor Letzelter, Nermin Samet, Renaud Marlet, Matthieu Cord, Patrick Perez, Eduardo Valle.
38th Annual Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, CA, December 2024.
- A Study of Test-time Contrastive Concepts for Open-world, Open-vocabulary Semantic Segmentation.
Monika Wysoczańska, Antonin Vobecky, Amaia Cardiel, Tomasz Trzciński, Renaud Marlet, Andrei Bursuc, Oriane Siméoni.
Preprint arXiv:2407.05061, July 2024.
- Train Till You Drop: Towards Stable and Robust Source-free Unsupervised 3D Domain Adaptation.
Björn Michele, Alexandre Boulch, Tuan-Hung Vu, Gilles Puy, Renaud Marlet, Nicolas Courty.
35th European Conference on Computer Vision (ECCV 2024), Milan, Italy, Sept-Oct 2024.
Code.
- Valeo4Cast: A Modular Approach to End-to-End Forecasting.
Yihong Xu, Éloi Zablocki, Alexandre Boulch, Gilles Puy, Mickael Chen, Florent Bartoccioni, Nermin Samet, Oriane Siméoni, Spyros Gidaris, Tuan-Hung Vu, Andrei Bursuc, Eduardo Valle, Renaud Marlet, Matthieu Cord.
Winning solution of the Argoverse 2 "Unified Detection, Tracking, and Forecasting" challenge (a.k.a. "End-to-End Forecasting Challenge"), held at the CVPR 2024 Workshop on Autonomous Driving (WAD 2024), Seattle, WA, USA, June 2024.
Revised version: Modular4Cast: A Modular Approach to End-to-End Forecasting accepted at the ECCV 2024 3rd Workshop on Event Detection for Situation Awareness in Autonomous Driving (ROAD++ 2024)
- OccFeat: Self-supervised Occupancy Feature Prediction for Pretraining BEV Segmentation Networks.
Sophia Sirko-Galouchenko, Alexandre Boulch, Spyros Gidaris, Andrei Bursuc, Antonin Vobecky, Patrick Pérez, Renaud Marlet.
CVPR 2024 Workshop on Autonomous Driving (WAD 2024), Seattle, WA, USA, June 2024.
Code.
- Three Pillars Improving Vision Foundation Model Distillation for Lidar.
Gilles Puy, Spyros Gidaris, Alexandre Boulch, Oriane Siméoni, Corentin Sautier, Patrick Pérez, Andrei Bursuc, Renaud Marlet.
37th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2024), Seattle, WA, USA, June 2024.
Code.
- NOPE: Novel Object Pose Estimation from a Single Image.
Van Nguyen Nguyen, Thibault Groueix, Georgy Ponimatkin, Yinlin Hu, Renaud Marlet, Mathieu Salzmann, Vincent Lepetit.
37th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2024), Seattle, WA, USA, June 2024.
Project, code.
- SALUDA: Surface-based Automotive Lidar Unsupervised Domain Adaptation.
Björn Michele, Alexandre Boulch, Gilles Puy, Tuan-Hung Vu, Renaud Marlet, Nicolas Courty.
International Conference on 3D Vision (3DV 2024) [spotlight], Davos, Switzerland, March 2024.
Code.
- BEVContrast: Self-Supervision in BEV Space for Automotive Lidar Point Clouds.
Corentin Sautier, Gilles Puy, Alexandre Boulch, Vincent Lepetit, Renaud Marlet.
International Conference on 3D Vision (3DV 2024), Davos, Switzerland, March 2024.
Code.
- DiffHPE: Robust, Coherent 3D Human Pose Lifting with Diffusion.
Cédric Rommel, Eduardo Valle, Mickaël Chen, Souhaiel Khalfaoui, Renaud Marlet, Matthieu Cord, Patrick Pérez.
11th IEEE International Workshop on Analysis and Modeling of Faces and Gestures (AMFG 2023) in conjunction with ICCV 2023, Paris, France, October 2023.
Code.
- You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentation.
Nermin Samet, Oriane Siméoni, Gilles Puy, Georgy Ponimatkin, Renaud Marlet, Vincent Lepetit.
19th CVF/IEEE International Conference on Computer Vision (ICCV 2023), Paris, France, October 2023.
Code.
- Using a Waffle Iron for Automotive Point Cloud Semantic Segmentation.
Gilles Puy, Alexandre Boulch, Renaud Marlet.
19th CVF/IEEE International Conference on Computer Vision (ICCV 2023), Paris, France, October 2023.
Code.
- RangeViT: Towards Vision Transformers for 3D Semantic Segmentation in Autonomous Driving.
Angelika Ando, Spyros Gidaris, Andrei Bursuc, Gilles Puy, Alexandre Boulch, Renaud Marlet.
36th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2023), Vancouver, Canada, June 2023.
Code.
- ALSO: Automotive Lidar Self-supervision by Occupancy estimation.
Alexandre Boulch, Corentin Sautier, Björn Michele, Gilles Puy, Renaud Marlet.
36th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2023), Vancouver, Canada, June 2023.
Code.
- Few-Shot Object Detection and Viewpoint Estimation for Objects in The Wild.
Yang Xiao, Vincent Lepetit, Renaud Marlet.
IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI 2023), 45(3), March 2023.
Extended version of ECCV 2020 paper with the same title.
Project page with code.
- A Simple and Powerful Global Optimization for Unsupervised Video Object Segmentation.
Georgy Ponimatkin, Nermin Samet, Yang Xiao, Yuming Du, Renaud Marlet, Vincent Lepetit
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2023), Waikoloa, HA, USA, January 2023.
Project page and code.
- Deep Surface Reconstruction from Point Clouds with Visibility Information.
Raphael Sulzer, Loic Landrieu, Alexandre Boulch, Renaud Marlet, Bruno Vallet.
26th International Conference on Pattern Recognition (ICPR 2022), Montreal, Quebec, Canada, August 2022.
Code and data.
- VASAD: a Volume and Semantic dataset for Building Reconstruction from Point Clouds.
Pierre-Alain Langlois, Yang Xiao, Alexandre Boulch, Renaud Marlet.
26th International Conference on Pattern Recognition (ICPR 2022), Montreal, Quebec, Canada, August 2022.
Code and data, supplementary material.
- POCO: Point Convolution for Surface Reconstruction.
Alexandre Boulch, Renaud Marlet.
35th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2022), New Orleans, LO, USA, June 2022.
Code.
- Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data.
Corentin Sautier, Gilles Puy, Spyros Gidaris, Alexandre Boulch, Andrei Bursuc, Renaud Marlet.
35th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2022), New Orleans, LO, USA, June 2022.
Code of SlidR, supplementary material.
- Spherical Perspective on Learning with Normalization Layers.
Simon Roburin, Yann de Mont-Marin, Andrei Bursuc, Renaud Marlet, Patrick Pérez, Mathieu Aubry.
Neurocomputing, 487:66-74, May 2022.
Short version, titled "A spherical analysis of Adam with Batch Normalization",
at International Workshop on Optimization for Machine Learning (OPT 2021), satellite event of NeurIPS 2021.
Project page (with code).
- PoseContrast: Class-Agnostic Object Viewpoint Estimation in the Wild with Pose-Aware Contrastive Learning.
Yang Xiao, Yuming Du, Renaud Marlet.
International Conference on 3D Vision (3DV 2021) [oral], virtual, December 2021.
Project page with code.
- Generative Zero-Shot Learning for Semantic Segmentation of 3D Point Clouds.
Björn Michele, Alexandre Boulch, Gilles Puy, Maxime Bucher, Renaud Marlet.
International Conference on 3D Vision (3DV 2021), virtual, December 2021.
- NeeDrop: Self-supervised Shape Representation from Sparse Point Clouds using Needle Dropping.
Alexandre Boulch, Pierre-Alain Langlois, Gilles Puy, Renaud Marlet.
International Conference on 3D Vision (3DV 2021), virtual, December 2021.
Supplementary material.
- Localizing Objects with Self-Supervised Transformers and no Labels.
Oriane Siméoni, Gilles Puy, Huy V. Vo, Simon Roburin, Spyros Gidaris, Andrei Bursuc, Patrick Pérez, Renaud Marlet, Jean Ponce.
32th British Machine Vision Conference (BMVC 2021), virtual, November 2021.
Project page with code.
- PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds.
Anh-Quan Cao, Gilles Puy, Alexandre Boulch, Renaud Marlet.
18th CVF/IEEE International Conference on Computer Vision (ICCV 2021), Virtual/Online, September 2021.
Code and supplementary material.
- 3D Reconstruction by Parameterized Surface Mapping.
Pierre-Alain Langlois, Matthew Fisher, Oliver Wang, Vladimir Kim, Alexandre Boulch, Renaud Marlet, Bryan Russell.
28th IEEE International Conference on Image Processing (ICIP 2021), Anchorage, Alaska, September 2021.
Supplementary material.
- Scalable Surface Reconstruction with Delaunay-Graph Neural Networks.
Raphael Sulzer, Loïc Landrieu, Renaud Marlet, Bruno Vallet.
Computer Graphics Forum (CGF 2021), 40(5):157-167, August 2021.
19th Eurographics Symposium on Geometry Processing (SGP 2021), Toronto, Ontario, online, July 2021.
- FKAConv: Feature-Kernel Alignment for Point Cloud Convolution.
Alexandre Boulch, Gilles Puy, Renaud Marlet.
15th Asian Conference on Computer Vision (ACCV 2020) [oral], online, November 2020.
Projet page with code.
- Pixel-Pair Occlusion Relationship Map (P2ORM): Formulation, Inference & Application.
Xuchong Qiu, Yang Xiao, Chaohui Wang, Renaud Marlet.
16th European Conference on Computer Vision (ECCV 2020) [spotlight], online, August 2020.
Project page, code and data, supplementary material.
- Few-Shot Object Detection and Viewpoint Estimation for Objects in The Wild.
Yang Xiao, Renaud Marlet.
16th European Conference on Computer Vision (ECCV 2020), online, August 2020.
Project page with code, slides and videos.
- FLOT: Scene Flow On Point Clouds Guided by Optimal Transport.
Gilles Puy, Alexandre Boulch, and Renaud Marlet.
16th European Conference on Computer Vision (ECCV 2020), online, August 2020.
Project page with code and data.
- Approximating Shapes in Images with Low-Complexity Polygons.
Muxingzi Li, Florent Lafarge, Renaud Marlet.
33rd IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2020) [oral], Seattle, WA, USA, June 2020.
Code.
- Surface Reconstruction from 3D Line Segments.
Pierre-Alain Langlois, Alexandre Boulch, Renaud Marlet.
International Conference on 3D Vision (3DV 2019) [oral], Québec City, Canada, September 2019.
Project page, supplementary material (text, video), code.
- Pose from Shape: Deep Pose Estimation for Arbitrary 3D Objects.
Yang Xiao, Xuchong Qiu, Pierre-Alain Langlois, Mathieu Aubry, Renaud Marlet.
30th British Machine Vision Conference (BMVC 2019), Cardiff, United Kingdom, September 2019.
Supplementary material,
project page (code and data).
- Virtual Training for a Real Application: Accurate Object-Robot Relative Localization without Calibration.
Vianney Loing, Renaud Marlet, Mathieu Aubry.
International Journal of Computer Vision (IJCV 2018), 126(9), pp 1045-1060, September 2018.
Project page (dataset, code and video).
- Efficient 2D and 3D Facade Segmentation using Auto-Context.
Raghudeep Gadde, Varun Jampani, Renaud Marlet, Peter Gehler.
IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI 2018), 40(5), May 2018.
Supplementary material.
- Line-based Robust SfM with Little Image Overlap.
Yohann Salaün, Renaud Marlet, Pascal Monasse.
5th International Conference on 3D Vision (3DV 2017), Qingdao, China, October 2017.
Supplementary material, code (LineSfM).
- Patchwork Stereo: Scalable, Structure-aware 3D Reconstruction in Man-made Environments.
Amine Bourki, Martin de La Gorce, Renaud Marlet, Nikos Komodakis.
IEEE Winter Conference on Applications of Computer Vision (WACV 2017), Santa Rosa, CA, USA, March 2017.
Supplementary material.
- Deep Learning for Urban Remote Sensing.
Nicolas Audebert, Alexandre Boulch, Hicham Randrianarivo, Bertrand Le Saux, Marin Ferecatu, Sébastien Lefèvre, Renaud Marlet.
Joint Urban Remote Sensing Event (JURSE 2017) [invited paper], Dubai, UEA, March 2017.
- OpenMVG: Open Multiple View Geometry.
Pierre Moulon, Pascal Monasse, Romuald Perrot, Renaud Marlet.
1st Workshop on Reproducible Research in Pattern Recognition (RRPR 2016), Cancun, Mexico, December 2016.
Code (OpenMVG).
- The multiscale line segment detector.
Yohann Salaün, Renaud Marlet, Pascal Monasse.
1st Workshop on Reproducible Research in Pattern Recognition (RRPR 2016), Cancun, Mexico, December 2016.
Code (MLSD).
- Multiscale line segment detector for robust and accurate SfM.
Yohann Salaün, Renaud Marlet, Pascal Monasse.
23rd International Conference on Pattern Recognition (ICPR 2016), Cancun, Mexico, December 2016.
Code (MLSD).
- Robust and accurate line- and/or point-based pose estimation without Manhattan assumptions.
Yohann Salaün, Renaud Marlet, Pascal Monasse.
14th European Conference on Computer Vision (ECCV 2016), Amsterdam, The Netherlands, October 2016.
Supplementary material, code (LineSfM).
- Crafting a multi-task CNN for viewpoint estimation.
Francisco Massa, Renaud Marlet, Mathieu Aubry.
27th British Machine Vision Conference (BMVC 2016), York, United Kingdom, September 2016.
- Deep Learning for Robust Normal Estimation in Unstructured Point Clouds.
Alexandre Boulch, Renaud Marlet.
Computer Graphics Forum (CGF 2016), 35(5), August 2016.
14th Eurographics Symposium on Geometry Processing (SGP 2016), Berlin, Germany, June 2016.
Slides, code.
- Learning grammars for architecture-specific facade parsing.
Raghudeep Gadde, Renaud Marlet, Nikos Paragios.
International Journal of Computer Vision (IJCV 2016), 117(3), 290-316, March 2016.
Dataset.
- A MRF Shape Prior for Facade Parsing with Occlusions.
Mateusz Koziński, Raghudeep Gadde, Sergey Zagoruyko, Renaud Marlet, Guillaume Obozinski.
28th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2015), Boston, June 2015.
Supplementary material.
- Beyond procedural facade parsing: Bidirectional alignment via linear programming.
Mateusz Koziński, Guillaume Obozinski, Renaud Marlet.
12th Asian Conference on Computer Vision (ACCV 2014), Singapore, November 2014.
Supplementary material.
- Match Selection and Refinement for Highly Accurate Two-View Structure from Motion.
Zhe Liu, Pascal Monasse, Renaud Marlet.
13rd European Conference on Computer Vision (ECCV 2014) [oral], Zurich, Switzerland, September 2014.
Supplementary material, code.
- Statistical criteria for shape fusion and selection.
Alexandre Boulch, Renaud Marlet.
22nd International Conference on Pattern Recognition (ICPR 2014), Stockholm, Sweden, August 2014.
Code.
- Piecewise-Planar 3D Reconstruction with Edge and Corner Regularization.
Alexandre Boulch, Martin de La Gorce, Renaud Marlet.
Computer Graphics Forum (CGF 2014), 33(5), 55-64, August 2014.
12th Eurographics Symposium on Geometry Processing (SGP 2014), Cardiff, UK, July 2014.
Slides.
- Image Parsing with Graph Grammars and Markov Random Fields.
Mateusz Koziński, Renaud Marlet.
IEEE Winter Conference on Applications of Computer Vision (WACV 2014), Steamboat Springs, CO, USA, March 2014.
- Global Fusion of Relative Motions for Robust, Accurate and Scalable Structure from Motion.
Pierre Moulon, Pascal Monasse, Renaud Marlet.
14th IEEE International Conference on Computer Vision (ICCV 2013), Sydney, Australia, December 2013.
Code (OpenMVG).
- Semantizing Complex 3D Scenes using Constrained Attribute Grammars.
Alexandre Boulch, Simon Houllier, Renaud Marlet, Olivier Tournaire.
Computer Graphics Forum (CGF 2013), 32(5), 33-42, August 2013.
11th Eurographics Symposium on Geometry Processing (SGP 2013), Genoa, Italy, July 2013.
Supplementary material, slides.
- Efficient and Scalable 4th-order Match Propagation.
David Ok, Renaud Marlet, Jean-Yves Audibert.
11th Asian Conference on Computer Vision (ACCV 2012), Daejeon, Corea, November 2012.
Supplementary material.
- Adaptive Structure from Motion with a contrario model estimation.
Pierre Moulon, Pascal Monasse, Renaud Marlet.
11th Asian Conference on Computer Vision (ACCV 2012), Daejeon, Corea, November 2012.
Code (OpenMVG).
- High-Level Bottom-Up Cues for Top-Down Parsing of Facade Images.
David Ok, Mateusz Koziński, Renaud Marlet, Nikos Paragios.
2nd Joint 3DIM/3DPVT Conference on 3D Imaging, Modeling, Processing, Visualization and Transmission (3DIMPVT 2012), Zürich, Switzerland, October 2012.
- Virtual Line Descriptor and Semi-Local Graph Matching Method for Reliable Feature Correspondence.
Zhe Liu, Renaud Marlet.
23rd British Machine Vision Conference (BMVC 2012), Surrey, United Kingdom, September 2012.
Supplementary material, code (KVLD).
- Fast and Robust Normal Estimation for Point Clouds with Sharp Features.
Alexandre Boulch, Renaud Marlet.
Computer Graphics Forum (CGF 2012), 31(5), 1765-1774, August 2012.
10th Eurographics Symposium on Geometry Processing (SGP 2012), Tallinn, Estonia, July 2012.
Supplementary material, slides, code.
- Indoor Calibration using Segments Chains.
Jamil Draréni, Renaud Keriven, Renaud Marlet.
33rd Annual Symposium of the German Association for Pattern Recognition (GCPR/DAGM 2011), Frankfurt, Germany, August 2011.
You may also take a look at my publications in other lives,
concerning natural language processing
(NLP) as well as programming
languages, software engineering and security.
Books
The publisher had these
two volumes transmogrified
into some sort of
technical English
|
|
and bundled into a single book
titled Program Specialization.
|
Please refer to the French version if you come across inconsistencies in the English one.
Teaching [ Enseignement ]
Vision 3D artificielle [ENS Paris Saclay, Master MVA]
Vision algorithmique : reconstruction 3D (VIALG) [UPMC/Télécom, Master d'informatique, spécialité IMA]
Resume