Multi-Task Deep Learning for Satellite Image Pansharpening and Segmentation

Andrew Khalel (1, 2) , Onur Tasar (1) , Guillaume Charpiat (3) , Yuliya Tarabalka (1)

1 TITANE - Geometric Modeling of 3D Environments
2 Cairo University
3 TAU - TAckling the Underspecified

In this work, we propose a novel multi-task framework, to learn satellite image pansharpening and segmentation jointly. Our framework is based on the encoder-decoder architecture, where both tasks share the same encoder but each one has its own decoder. We compare our framework against single-task models with different architectures. Results show that our framework outperforms all other approaches in both tasks.

Keywords: Joint learning; Segmentation; Multi-task; Pansharpening

Domains: Artificial Intelligence [cs.AI]; Computer Vision and Pattern Recognition [cs.CV]; Machine Learning [cs.LG]; Signal and Image Processing

Andrew Khalel, Onur Tasar, Guillaume Charpiat, Yuliya Tarabalka. Multi-Task Deep Learning for Satellite Image Pansharpening and Segmentation. IGARSS 2019 - IEEE International Geoscience and Remote Sensing Symposium, Jul 2019, Yokohama, Japan. pp.4869-4872, ⟨10.1109/IGARSS.2019.8899851⟩. ⟨hal-02276549⟩