We created NPM3D Benchmark Suite on 3D Point Clouds: http://npm3d.fr/
You can contact jean-emmanuel.deschaud@mines-paristech.fr or francois.goulette@mines-paristech.fr
Research Team NPM3D
Research areas of the NMP3D team are around 3D data as “point cloud”:
– Mobile LiDAR systems
– Point cloud rendering (screen-space rendering, splatting)
– LiDAR SLAM, Localization on Map
– HD Maps for autonomous vehicles
– Point Cloud processings from archeological sites / objects
– Object recognition / scene segmentation and classification in LiDAR data
– Deep Learning for 3D Point Clouds
We have multiple 3D sensors (Kinect 2, scanner FARO Focus X130, Velodyne VLP16, HDL32) and two mobile mapping platforms:
– L3D2: vehicle with GPS/IMU localization system, Velodyne HDL32 and Ladybug5
– Drone with Velodyne VLP16 and Camera FLIR Blackfly
People
• Researchers: – François Goulette: Full Professor, francois.goulette@mines-paristech.fr
– Jean-Emmanuel Deschaud: Associate Professor, jean-emmanuel.deschaud@mines-paristech.fr
• PhD students:
– Jules Sanchez (2020-): Real-time semantic segmentation of lidar data for the autonomous vehicle
– Pierre Dellenbach (2020-): Self-supervised Deep SLAM on camera and LiDAR data
– Jean-Pierre Richa (2019-): Point Cloud rendering through splatting for LiDAR Simulation
– Sofiane Horache (2019-): Automatic recognition of patterns on 3D curved surfaces and its application to Celtic art
– David Duque (2018-): Segmentation and classification of road scenes by analysis of LIDAR data, color and multi-spectral images (in cooperation with CMM lab)
• Former PhD students:
– Hugues Thomas (2016-2019): Learning new representations for 3D point cloud semantization (now Post-Doc at University of Toronto)
– Xavier Roynard (2015-2019): Semantization of 3D Points Clouds acquired by Embedded System (working now at SAFRAN as R&D engineer)
– Hassan Bouchiba (2014-2018): Contributions in point-based processing for rendering and simulation in fluid mechanics (working now at Terra3D)
– Houssem Nouira (2013-2016): Point cloud refinement with self-calibration of a mobile multi-beam lidar (working now at MENSI/TRIMBLE)
Selected Papers
Automatic clustering of celtic coins based on 3D Point Cloud Pattern Analysis, ISPRS Congress, 2020, pdf, video
SHREC 2020 Track: 3D Point Cloud Semantic Segmentation for Street Scenes, Computer & Graphics, 2020, pdf
Computational Fluid Dynamics on 3D Point Set Surfaces, Journal of Computational Physics, 2020, pdf
KPConv: Flexible and Deformable Convolution for Point Clouds, ICCV, 2019, pdf, code
A Graph-based Color Lines model for image analysis, ICIAP, 2019, pdf
Paris-Lille-3D: A large and high-quality ground-truth urban point cloud dataset for automatic segmentation and classification, IJRR, 2018, pdf
Classification of Point Cloud for Road Scene Understanding with Multiscale Voxel Deep Network, Workshop IROS, 2018, pdf, code
Semantic Classification of 3D Point Clouds with Multiscale Spherical Neighborhoods, 3DV, 2018, pdf
Raw point cloud deferred shading through screen space pyramidal operators, Short Paper EUROGRAPHICS, 2018, video, pdf
IMLS-SLAM: scan-to-model matching based on 3D data, ICRA, 2018, pdf
Point Cloud Refinement with Self-calibration of a mobile multi-beam Lidar Sensor, The Photogrammetric Record, 2017, pdf
Current research projects
REPLICA : Massive computing platform for autonomous vehicle with realistic sensors simulation
ARCHEO 3D : Automatic recognition of shapes and more particularly carved patterns repeated by stamping on different celtic scanned object
Teaching
• Course on 3D Point Clouds in the Master 2 MVA at ENS Paris-Saclay and Master 2 IASD at PSL University: course webpage here
If you do not know “3D Point Cloud data”
Point Cloud of MINES ParisTech (FARO Focus X130)
Point Cloud of MINES ParisTech (FARO Focus X130)
Point Cloud of the city of Lille (our plateform L3D2 with Velodyne HDL32)