Abstract | ||
---|---|---|
Deep learning models for semantic segmentation rely on expensive, large-scale, manually annotated datasets. Labelling is a tedious process that can take hours per image. Automatically annotating video sequences by propagating sparsely labeled frames through time is a more scalable alternative. In this work, we propose a novel label propagation method, termed Warp-Refine Propagation, that combines semantic cues with geometric cues to efficiently auto-label videos. Our method learns to refine geometrically-warped labels and infuse them with learned semantic priors in a semi-supervised setting by leveraging cycle consistency across time. We quantitatively show that our method improves label-propagation by a noteworthy margin of 13.1 mIoU on the ApolloScape dataset. Furthermore, by training with the auto-labelled frames, we achieve competitive results on three semantic-segmentation benchmarks, improving the state-of-the-art by a large margin of 1.8 and 3.61 mIoU on NYU-V2 and KITTI, while matching the current best results on Cityscapes. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/ICCV48922.2021.01521 | ICCV |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Aditya Ganeshan | 1 | 1 | 1.02 |
Alexis Vallet | 2 | 0 | 0.68 |
Yasunori Kudo | 3 | 0 | 0.68 |
Shin-ichi Maeda | 4 | 26 | 8.11 |
Tommi Kerola | 5 | 35 | 4.13 |
Rares Ambrus | 6 | 76 | 11.59 |
Dennis Park | 7 | 0 | 0.34 |
Adrien Gaidon | 8 | 0 | 0.34 |