Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling (Invited Paper)
This addresses urban mapping for remote sensing applications, but appears incremental as it builds on existing dual-stream architectures.
The paper tackles urban semantic labeling by fusing heterogeneous DSM and IRRG optical data using a novel residual correction module in fully convolutional networks, achieving new state-of-the-art results on the ISPRS Vaihingen dataset.
In this work, we present a novel module to perform fusion of heterogeneous data using fully convolutional networks for semantic labeling. We introduce residual correction as a way to learn how to fuse predictions coming out of a dual stream architecture. Especially, we perform fusion of DSM and IRRG optical data on the ISPRS Vaihingen dataset over a urban area and obtain new state-of-the-art results.