Fully convolutional networks for street furniture identification in panorama images

Fully convolutional networks for street furniture identification in panorama images

Ao, Y. (University of Twente)
Wang, J. (TU Delft Optical and Laser Remote Sensing)
Zhou, M. (Chinese Academy of Sciences)
Lindenbergh, R.C. (TU Delft Optical and Laser Remote Sensing)
Yang, M. Y. (University of Twente)

2019

Panoramic images are widely used in many scenes, especially in virtual reality and street view capture. However, they are new for street furniture identification which is usually based on mobile laser scanning point cloud data or conventional 2D images. This study proposes to perform semantic segmentation on panoramic images and transformed images to separate light poles and traffic signs from background implemented by pre-trained Fully Convolutional Networks (FCN). FCN is the most important model for deep learning applied on semantic segmentation for its end to end training process and pixel-wise prediction. In this study, we use FCN-8s model that pre-trained on cityscape dataset and finetune it by our own data. The results show that in both pre-trained model and fine-tuning, transformed images have better prediction results than panoramic images.

Fully Convolutional Networks
Object Identification
Panoramic Images
Semantic Segmentation
Street Furniture

http://resolver.tudelft.nl/uuid:1df54921-b164-4d4d-a796-cd1c910049ea

https://doi.org/10.5194/isprs-archives-XLII-2-W13-13-2019

1682-1750

International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLII (2/W13), 13-20

4th ISPRS Geospatial Week 2019, 2019-06-10 → 2019-06-14, Enschede, Netherlands

Institutional Repository

journal article

© 2019 Y. Ao, J. Wang, M. Zhou, R.C. Lindenbergh, M. Y. Yang