Fusion of heterogeneous data in convolutional networks for urban semantic labeling

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.

• Publication year

  2017

• Venue

  Joint Urban Remote Sensing Event

• Publication date

  2017-01-20

• Fields of study

  Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1109/JURSE.2017.7924566arXiv 1701.05818
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• SEMANTIC SEGMENTATION OF AERIAL IMAGES WITH AN ENSEMBLE OF CNNS

  2016cited by this paper
• Spatiotemporal scene interpretation of space videos via deep neural network and tracklet analysis

  2016cited by this paper
• Deep Learning Earth Observation Classification Using ImageNet Pretrained Networks

  2016cited by this paper
• An Efficient Framework for Pixel-wise Building Segmentation from Aerial Images

  2015cited by this paper
• Multimodal deep learning for robust RGB-D object recognition

  2015cited by this paper
• Benchmarking classification of earth-observation data: From learning explicit features to convolutional networks

  2015cited by this paper
• Effective semantic pixel labelling with convolutional networks and Conditional Random Fields

  2015cited by this paper
• Deep Residual Learning for Image Recognition

  2015influential reference
• SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation

  2015influential reference
• The Pascal Visual Object Classes Challenge: A Retrospective

  2014cited by this paper
• Fully convolutional networks for semantic segmentation

  2014cited by this paper
• Use of the stair vision library within the ISPRS 2D semantic labeling benchmark (Vaihingen)

  2014cited by this paper
• The ISPRS benchmark on urban object classification and 3D building reconstruction

  2012cited by this paper
• Multimodal Deep Learning

  2011cited by this paper
• The DGPF-Test on Digital Airborne Camera Evaluation - Over- view and Test Design

  2010cited by this paper
• International Conference on Intelligent Robots and Systems

  year unknowncited by this paper

• Estimating vegetation indices and biophysical parameters for Central European temperate forests with Sentinel-1 SAR data and machine learning

  2023cites this paper
• Hazardous atmospheric dispersion in urban areas: A Deep Learning approach for emergency pollution forecast

  2022cites this paper
• A Multi-Modal Fusion Network Guided by Feature Co-Occurrence for Urban Region Function Recognition

  2022cites this paper
• Semantic Segmentation of Satellite Images Using Deep-Unet

  2022cites this paper
• HFCER : Hybrid Fusion for Cultural Event Recognition in Images

  2021cites this paper
• A Mixed Method For Internal Threat Detection

  2021cites this paper
• Semantically Enriched Crop Type Classification and Linked Earth Observation Data to Support the Common Agricultural Policy Monitoring

  2021cites this paper
• Semantic Web and IoT

  2021cites this paper
• A Survey of Semantic Construction and Application of Satellite Remote Sensing Images and Data

  2021cites this paper
• A real-time efficient object segmentation system based on U-Net using aerial drone images

  2021cites this paper
• INTELLIGENT MAPPING OF IRRIGATED AREAS FROM LANDSAT 8 IMAGES USING TRANSFER LEARNING

  2021cites this paper
• Team OBELIX IRISA Activity Report 2017 1 Team composition Head of the team

  2020influential citation
• Semantic Segmentation of Natural Materials on a Point Cloud Using Spatial and Multispectral Features

  2020cites this paper
• Height Estimation From Single Aerial Images Using a Deep Ordinal Regression Network

  2020cites this paper
• Small Object Detection in Highly Variable Backgrounds

  2019cites this paper
• A Survey on Mobile Crowd-Sensing and Its Applications in the IoT Era

  2019cites this paper
• Deep Learning Approach on Information Diffusion in Heterogeneous Networks

  2019cites this paper
• Injecting spatial priors in Earth observation with machine vision

  2019cites this paper
• Correspondence matching in unorganized 3D point clouds using Convolutional Neural Networks

  2019cites this paper
• Decision Fusion of Remote-Sensing Data for Land Cover Classification

  2019cites this paper
• Novelty Detection in Social Media by Fusing Text and Image Into a Single Structure

  2019cites this paper
• Traitement d'images à haute résolution grâce à des techniques d'apprentissage en profondeur. (Processing high-resolution images through deep learning techniques)

  2018cites this paper
• IM2HEIGHT: Height Estimation from Single Monocular Imagery via Fully Residual Convolutional-Deconvolutional Network

  2018cites this paper
• Land cover mapping at very high resolution with rotation equivariant CNNs: towards small yet accurate models

  2018cites this paper
• Vehicle Instance Segmentation From Aerial Image and Video Using a Multitask Learning Residual Fully Convolutional Network

  2018cites this paper
• Feature Fusion Through Multitask CNN for Large-scale Remote Sensing Image Segmentation

  2018cites this paper
• Convolutional Neural Networks on Multiple Respiratory Channels to Detect Hypopnea and Obstructive Apnea Events

  2018cites this paper
• Effective Building Extraction by Learning to Detect and Correct Erroneous Labels in Segmentation Mask

  2018cites this paper
• Improved convolutional neural network combined with rough set theory for data aggregation algorithm

  2018cites this paper
• A 3D convolutional neural network method for land cover classification using LiDAR and multi-temporal Landsat imagery

  2018cites this paper
• Analysis of Encoder-Decoder Based Deep Learning Architectures for Semantic Segmentation in Remote Sensing Images

  2018cites this paper
• Joint Learning from Earth Observation and OpenStreetMap Data to Get Faster Better Semantic Maps

  2017cites this paper
• Urban Land Cover Classification With Missing Data Modalities Using Deep Convolutional Neural Networks

  2017cites this paper
• Double-stream Convolutional Neural Networks for Machine Vision Inspection of Natural Products

  2017cites this paper
• Deep learning for semantic segmentation of remote sensing images with rich spectral content

  2017cites this paper