Loic Landrieu

Loic Landrieu
Institut national de l’information géographique et forestière | IGN · LaSTIG lab

PhD

About

58
Publications
11,937
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,383
Citations
Introduction
Hi! I am Loic Landrieu, currently employed as a full-time research scientist at the LaSTIG lab at the French Mapping Agency. I work at the interface between machine learning / computer vision and geospatial learning. In particular, I aim to exploit data structure to develop faster and more parsimonious algorithms. loiclandrieu.com -> PI of the READY3D ANR on real-time 3D perception -> Program Chair of the XXIV ISPRS Congress -> Co-Organizer of EarthVision CVPR Workshop

Publications

Publications (58)
Preprint
Full-text available
Machine learning techniques have proved useful for classifying and analyzing audio content. However, recent methods typically rely on abstract and high-dimensional representations that are difficult to interpret. Inspired by transformation-invariant approaches developed for image and 3D data, we propose an audio identification model based on learna...
Article
Full-text available
We propose a new deep learning-based method for estimating the occupancy of vegetation strata from airborne 3D LiDAR point clouds. Our model predicts rasterized occupancy maps for three vegetation strata corresponding to lower, medium, and higher cover. Our weakly-supervised training scheme allows our network to only be supervised with vegetation o...
Preprint
Full-text available
Roof-mounted spinning LiDAR sensors are widely used by autonomous vehicles, driving the need for real-time processing of 3D point sequences. However, most LiDAR semantic segmentation datasets and algorithms split these acquisitions into $360^\circ$ frames, leading to acquisition latency that is incompatible with realistic real-time applications and...
Article
Full-text available
Optical and radar satellite time series are synergetic: optical images contain rich spectral information, while C-band radar captures useful geometrical information and is immune to cloud cover. Motivated by the recent success of temporal attention-based methods across multiple crop mapping tasks, we propose to investigate how these models can be a...
Preprint
The analysis of the multi-layer structure of wild forests is an important challenge of automated large-scale forestry. While modern aerial LiDARs offer geometric information across all vegetation layers, most datasets and methods focus only on the segmentation and reconstruction of the top of canopy. We release WildForest3D, which consists of 29 st...
Preprint
Full-text available
Recent works on 3D semantic segmentation propose to exploit the synergy between images and point clouds by processing each modality with a dedicated network and projecting learned 2D features onto 3D points. Merging large-scale point clouds and images raises several challenges, such as constructing a mapping between points and pixels, and aggregati...
Preprint
Full-text available
Most current neural networks for reconstructing surfaces from point clouds ignore sensor poses and only operate on raw point locations. Sensor visibility, however, holds meaningful information regarding space occupancy and surface orientation. In this paper, we present two simple ways to augment raw point clouds with visibility information, so it c...
Conference Paper
Full-text available
We propose a new deep learning-based method for estimating the occupancy of vegetation strata from 3D point clouds captured from an aerial platform. Our model predicts rasterized occupancy maps for three vegetation strata: lower, medium, and higher strata. Our training scheme allows our network to only being supervized with values aggregated over c...
Preprint
Full-text available
We propose a new deep learning-based method for estimating the occupancy of vegetation strata from 3D point clouds captured from an aerial platform. Our model predicts rasterized occupancy maps for three vegetation strata: lower, medium, and higher strata. Our training scheme allows our network to only being supervized with values aggregated over c...
Preprint
Full-text available
Optical and radar satellite time series are synergetic: optical images contain rich spectral information, while C-band radar captures useful geometrical information and is immune to cloud cover. Motivated by the recent success of temporal attention-based methods across multiple crop mapping tasks, we propose to investigate how these models can be a...
Article
Full-text available
While annual crop rotations play a crucial role for agricultural optimization, they have been largely ignored for automated crop type mapping. In this paper, we take advantage of the increasing quantity of annotated satellite data to propose to model simultaneously the inter- and intra-annual agricultural dynamics of yearly parcel classification wi...
Preprint
Full-text available
While annual crop rotations play a crucial role for agricultural optimization, they have been largely ignored for automated crop type mapping. In this paper, we take advantage of the increasing quantity of annotated satellite data to propose the first deep learning approach modeling simultaneously the inter- and intra-annual agricultural dynamics o...
Preprint
Full-text available
In this paper, we revisit the classical representation of 3D point clouds as linear shape models. Our key insight is to leverage deep learning to represent a collection of shapes as affine transformations of low-dimensional linear shape models. Each linear model is characterized by a shape prototype, a low-dimensional shape basis and two neural net...
Article
Full-text available
We introduce a novel learning-based, visibility-aware, surface reconstruction method for large-scale, defect-laden point clouds. Our approach can cope with the scale and variety of point cloud defects encountered in real-life Multi-View Stereo (MVS) acquisitions. Our method relies on a 3D Delaunay tetrahedralization whose cells are classified as in...
Preprint
Full-text available
Unprecedented access to multi-temporal satellite imagery has opened new perspectives for a variety of Earth observation tasks. Among them, pixel-precise panoptic segmentation of agricultural parcels has major economic and environmental implications. While researchers have explored this problem for single images, we argue that the complex temporal p...
Preprint
Full-text available
We introduce a novel learning-based, visibility-aware, surface reconstruction method for large-scale, defect-laden point clouds. Our approach can cope with the scale and variety of point cloud defects encountered in real-life Multi-View Stereo (MVS) acquisitions. Our method relies on a 3D Delaunay tetrahedralization whose cells are classified as in...
Conference Paper
Full-text available
The increasing accessibility and precision of Earth observation satellite data offers considerable opportunities for industrial and state actors alike. This calls however for efficient methods able to process time-series on a global scale. Building on recent work employing multi-headed self-attention mechanisms to classify remote sensing time seque...
Preprint
Full-text available
We introduce Torch-Points3D, an open-source framework designed to facilitate the use of deep networks on3D data. Its modular design, efficient implementation, and user-friendly interfaces make it a relevant tool for research and productization alike. Beyond multiple quality-of-life features, our goal is to standardize a higher level of transparency...
Preprint
Full-text available
Not all errors are created equal. This is especially true for many key machine learning applications. In the case of classification tasks, the severity of errors can be summarized under the form of a cost matrix, which assesses the gravity of confusing each pair of classes. When the target classes are organized into a hierarchical structure, this m...
Article
Full-text available
Leveraging the recent availability of accurate, frequent, and multimodal (radar and optical) Sentinel-1 and -2 acquisitions, this paper investigates the automation of land parcel identi- fication system (LPIS ) crop type classification. Our approach allows for the automatic integration of temporal knowledge, i.e., crop rotations using existing parc...
Preprint
Full-text available
The increasing accessibility and precision of Earth observation satellite data offers considerable opportunities for industrial and state actors alike. This calls however for efficient methods able to process time-series on a global scale. Building on recent work employing multi-headed self-attention mechanisms to classify remote sensing time seque...
Preprint
Full-text available
Satellite image time series, bolstered by their growing availability, are at the forefront of an extensive effort towards automated Earth monitoring by international institutions. In particular, large-scale control of agricultural parcels is an issue of major political and economic importance. In this regard, hybrid convolutional-recurrent neural a...
Article
Full-text available
We introduce a new method for the piecewise-planar approximation of 3D data, including point clouds and meshes. Our method is designed to operate on large datasets (e.g. millions of vertices) containing planar structures, which are very frequent in anthropic scenes. Our approach is also adaptive to the local geometric complexity of the input data....
Preprint
Full-text available
We present a fully-supervized method for learning to segment data structured by an adjacency graph. We introduce the graph-structured contrastive loss, a loss function structured by a ground truth segmentation. It promotes learning vertex embeddings which are homogeneous within desired segments, and have high contrast at their interface. Thus, comp...
Preprint
Full-text available
We present a parallel version of the cut-pursuit algorithm for minimizing functionals involving the graph total variation. We show that the decomposition of the iterate into constant connected components, which is at the center of this method, allows for the seamless parallelization of the otherwise costly graph-cut based refinement stage. We demon...
Conference Paper
Full-text available
We propose a new supervized learning framework for oversegmenting 3D point clouds into superpoints. We cast this problem as learning deep embeddings of the local geometry and radiometry of 3D points, such that the border of objects presents high contrasts. The embeddings are computed using a lightweight neural network operating on the points' local...
Preprint
Full-text available
We propose a new supervized learning framework for oversegmenting 3D point clouds into superpoints. We cast this problem as learning deep embeddings of the local geometry and radiometry of 3D points, such that the border of objects presents high contrasts. The embeddings are computed using a lightweight neural network operating on the points' local...
Conference Paper
Full-text available
We introduce a new method for the piecewise-planar approximation of 3D data, including point clouds and meshes. Our method is designed to operate on large datasets (e.g. millions of vertices) containing planar structures, which are very frequent in an-thropic scenes. Our approach is also adaptive to the local geometric complexity of the input data....
Preprint
Full-text available
In this article, we investigate several structured deep learning models for crop type classification on multi-spectral time series. In particular, our aim is to assess the respective importance of spatial and temporal structures in such data. With this objective, we consider several designs of convolutional, recurrent, and hybrid neural networks, a...
Preprint
Full-text available
In this article, we investigate several structured deep learning models for crop type classification on multi-spectral time series. In particular, our aim is to assess the respective importance of spatial and temporal structures in such data. With this objective, we consider several designs of convolutional, recurrent, and hybrid neural networks, a...
Conference Paper
Full-text available
Automatic analysis of Sentinel image time series is recommended for monitoring agricultural land use in Europe. To improve classification capacities, we propose a temporal structured classification combining Sentinel images and former vintages of the Land-Parcel Identification System. Inter-annual crop rotations are learned and combined with the sa...
Conference Paper
Full-text available
We present an extension of the cut-pursuit algorithm, introduced by Landrieu and Obozinski (2017), to the graph total-variation regularization of functions with a separable nondifferentiable part. We propose a modified algorithmic scheme as well as adapted proofs of convergence. We also present a heuristic approach for handling the cases in which t...
Conference Paper
Full-text available
We propose a novel deep learning-based framework to tackle the challenge of semantic segmentation of large-scale point clouds of millions of points. We argue that the organization of 3D point clouds can be efficiently captured by a structure called superpoint graph (SPG), derived from a partition of the scanned scene into geometrically homogeneous...
Article
Full-text available
We propose working set/greedy algorithms to efficiently solve problems penalized, respectively, by the total variation on a general weighted graph and its l0 counterpart the total level-set boundary size when the piecewise constant solutions have a small number of distinct level sets; this is typically the case when the total level-set boundary siz...
Article
Full-text available
In this paper, we introduce a mathematical framework for obtaining spatially smooth semantic labelings of 3D point clouds from a pointwise classification. We argue that structured regularization offers a more versatile alternative to the standard graphical model approach. Indeed, our framework allows us to choose between a wide range of fidelity fu...
Conference Paper
Full-text available
In this paper, we focus on the classification of lidar point cloud data acquired via mobile laser scanning, whereby the classification relies on a context model based on a Conditional Random Field (CRF). We present two approximate inference algorithms based on belief propagation, as well as a graph-cut-based approach not yet applied in this context...
Conference Paper
Full-text available
Nous traitons le problème de la classification sémantique de nuages de points 3D LIDAR pour les scènes urbaines à partir d'un jeu d'apprentissage limité. Nous introdui-sons un modèle de segmentation non paramétrique pour les scènes urbaines formées par des objets anthropiques de formes simples. Notre modèle segmente la scène en ré-gions géométrique...
Article
Full-text available
We consider the problem of the semantic classification of 3D LiDAR point clouds obtained from urban scenes when the training set is limited. We propose a non-parametric segmentation model for urban scenes composed of anthropic objects of simple shapes, partionning the scene into geometrically-homogeneous segments which size is determined by the loc...
Thesis
Full-text available
Modeling complex processes often involve a high number of variables with an intricate correlation structure. For example, many spatially-localized processes display spatial regularity, as variables corresponding to neighboring regions are more correlated than distant ones. The formalism of weighted graphs allows us to capture relationships between...
Conference Paper
Full-text available
We propose working-set/greedy algorithms to efficiently find the solutions to convex optimization problems penalized respectively by the total variation and the Mumford Shah boundary size. Our algorithms exploit the piecewise constant structure of the level-sets of the solutions by recursively splitting them using graph cuts. We obtain significant...
Article
Full-text available
We present a preconditioning of a generalized forward-backward splitting algorithm for finding a zero of a sum of maximally monotone operators $\sum_{i=1}^{n} A_i + B$ with $B$ cocoercive, involving only the computation of $B$ and of the resolvent of each $A_i$ separately. This allows in particular to minimize functionals of the form $\sum_{i=1}^n...
Conference Paper
Full-text available
This paper introduces an extension to undirected graphical models of the classical continuous time Markov chains. This model can be used to solve a transductive or unsupervised multi-class classification problem at each point of a network defined as a set of nodes connected by segments of different lengths. The classification is performed not only...

Network

Cited By

Projects

Projects (3)
Project
Investigating the use of simplicial complexes as a reconstruction structure for processing 3D LiDAR point clouds.
Project
Cut pursuit goal is to leverage the graph-regularity of he solution of common optimization problems involving graph-structured regularizers (total variation, contour length, etc...). Our method consists in maintaining and splitting iteratively a partition of the graph with graph cuts. For problems with regular solution, our methods offer speed-up of several orders of magnitude, with many applicative and improvement perspectives.
Project
Developing learning and optimization methods for fast and precise semantic segmentation of large-scale semantic segmentation of LiDAR 3D point clouds. Our approach consists in exploiting the structure of the point clouds into semantic objects to design adaptive graph-structured deep learning frameworks.