Training of Convolutional Networks on Multiple Heterogeneous Datasets for Street Scene Semantic Segmentation
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1045-1050
Abstract
We propose a convolutional network with hierarchical classifiers for per-pixel semantic segmentation, which is able to be trained on multiple, heterogeneous datasets and exploit their semantic hierarchy. Our network is the first to be simultaneously trained on three different datasets from the intelligent vehicles domain, i.e. Cityscapes, GTSDB and Mapillary Vistas, and is able to handle different semantic levelof-detail, class imbalances, and different annotation types, i.e. dense per-pixel and sparse bounding-box labels. We assess our hierarchical approach, by comparing against flat, nonhierarchical classifiers and we show improvements in mean pixel accuracy of 13.0% for Cityscapes classes and 2.4% for Vistas classes and 32.3% for GTSDB classes. Our implementation achieves inference rates of 17 fps at a resolution of 520 × 706 for 108 classes running on a GPU.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- The Mapillary Vistas Dataset for Semantic Understanding of Street Scenes2017
- Semi-automatic image annotation of street scenes2017
- A survey on deep learning-based fine-grained object classification and semantic segmentationInternational Journal of Automation and Computing, 2017
- Hierarchical Semantic Classification and Attribute Relations Analysis with Clothing Region DetectionLecture Notes in Electrical Engineering, 2016
- Deep Residual Learning for Image Recognition2016
- The Cityscapes Dataset for Semantic Urban Scene Understanding2016
- Weakly-and Semi-Supervised Learning of a Deep Convolutional Network for Semantic Image Segmentation2015
- Detection of traffic signs in real-world images: The German traffic sign detection benchmark2013