LaSTIG Seminar: Deep Learning for Environment Monitoring
Open virtual scientific seminar, 10 Nov 2021.
Combined advances in deep learning and Earth Observation technology have ushered in a new era of remote sensing. We are now able to efficiently oversee agricultural and forest lands at an unprecedented scale, and start to address pressing ecological challenges. This seminar comprises 6 talks by 3 international researchers and 3 students the LaSTIG Lab.
Chapters:
00:00 Intro
01:20 Charlotte Pelletier, IRISA, Univ. Bretagne Sud: Deep Learning for the Analysis of Remote Sensing Time Series
40:25 Prof. Jan Dirk Wegner, EcoVision, ETH Zurich: Towards High Carbon Stock Maps at Global Scale
1:12:32 Vivien Sainte Fare Garnot, IGN-LaSTIG-Univ. Gustave Eiffel: Panoptic Segmentation of Satellite Images
1:54:34 Marc Russwurm, Computational Science and Earth Observation Laboratory, EPFL: Data-Driven Vegetation Modeling and Tackling Representation Shift with Few-Shot Meta-Learning Methods
2:31:31 Ekaterina Kalinicheva, IGN, INRAE: Weakly Supervized Vegetation Stratum Analysis from Aerial LiDAR
2:51:46 Felix Quinton, IGN-LaSTIG-Univ. Gustave Eiffel: Modeling Crop Rotations with Deep Learning