Real-Time Semantic Segmentation with Label Propagation
Abstract
Despite of the success of convolutional neural networks for semantic image segmentation, CNNs cannot be used for many applications due to limited computational resources. Even efficient approaches based on random forests are not efficient enough for real-time performance in some cases. In this work, we propose an approach based on superpixels and label propagation that reduces the runtime of a random forest approach by factor 192 while increasing the segmentation accuracy.
Cite
Text
Sheikh et al. "Real-Time Semantic Segmentation with Label Propagation." European Conference on Computer Vision Workshops, 2016. doi:10.1007/978-3-319-48881-3_1Markdown
[Sheikh et al. "Real-Time Semantic Segmentation with Label Propagation." European Conference on Computer Vision Workshops, 2016.](https://mlanthology.org/eccvw/2016/sheikh2016eccvw-realtime/) doi:10.1007/978-3-319-48881-3_1BibTeX
@inproceedings{sheikh2016eccvw-realtime,
title = {{Real-Time Semantic Segmentation with Label Propagation}},
author = {Sheikh, Rasha and Garbade, Martin and Gall, Juergen},
booktitle = {European Conference on Computer Vision Workshops},
year = {2016},
pages = {3-14},
doi = {10.1007/978-3-319-48881-3_1},
url = {https://mlanthology.org/eccvw/2016/sheikh2016eccvw-realtime/}
}