Bai Y., Durand J.B., Vincent Grégoire, Forbes F. (2024). Semantic segmentation of sparse irregular point clouds for leaf/wood discrimination. In :
Oh A. (ed.), Naumann T. (ed.), Globerson A. (ed.), Saenko K. (ed.), Hardt M. (ed.), Levine S. (ed.). NIPS '23: Proceedings of the 37th International Conference on Neural Information Processing Systems. New York : Curran Associates, 48293-48313. International Conference on Neural Information Processing Systems, 37.
Titre du document
Semantic segmentation of sparse irregular point clouds for leaf/wood discrimination
Année de publication
2024
Type de document
Colloque
Auteurs
Bai Y., Durand J.B., Vincent Grégoire, Forbes F.
In
Oh A. (ed.), Naumann T. (ed.), Globerson A. (ed.), Saenko K. (ed.), Hardt M. (ed.), Levine S. (ed.) NIPS '23: Proceedings of the 37th International Conference on Neural Information Processing Systems
Source
New York : Curran Associates, 2024,
48293-48313
Colloque
International Conference on Neural Information Processing Systems, 37.
LiDAR (Light Detection And Ranging) has become an essential part of the remote sensing toolbox used for biosphere monitoring. In particular, LiDAR provides the opportunity to map forest leaf area with unprecedented accuracy, while leaf area has remained an important source of uncertainty affecting models of gas exchanges between the vegetation and the atmosphere. Unmanned Aerial Vehicles (UAV) are easy to mobilize and therefore allow frequent revisits, so as to track the response of vegetation to climate change. However, miniature sensors embarked on UAVs usually provide point clouds of limited density, which are further affected by a strong decrease in density from top to bottom of the canopy due to progressively stronger occlusion. In such a context, discriminating leaf points from wood points presents a significant challenge due in particular to strong class imbalance and spatially irregular sampling intensity. Here we introduce a neural network model based on the Pointnet ++ architecture which makes use of point geometry only (excluding any spectral information). To cope with local data sparsity, we propose an innovative sampling scheme which strives to preserve local important geometric information. We also propose a loss function adapted to the severe class imbalance. We show that our model outperforms state-of-the-art alternatives on UAV point clouds. We discuss future possible improvements, particularly regarding much denser point clouds acquired from below the canopy.
Plan de classement
Informatique, mathématiques, statistiques [040INFSTA]
;
Intelligence artificielle [122INTAR]
Localisation
Fonds IRD [F B010091540]
Identifiant IRD
fdi:010091540
