# Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling (Invited Paper)

## TL;DR

This paper introduces a novel convolutional network module that effectively fuses heterogeneous data sources, such as DSM and IRRG, for improved urban semantic labeling, achieving state-of-the-art results on the ISPRS Vaihingen dataset.

## Contribution

The paper proposes a residual correction-based fusion module for dual stream convolutional networks, enhancing urban semantic labeling performance.

## Key findings

- Achieved state-of-the-art accuracy on ISPRS Vaihingen dataset
- Demonstrated effective fusion of DSM and IRRG data
- Improved semantic labeling results over existing methods

## Abstract

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.