Submission details
Method name | PFSL FGSN 200 classes (sequential localization) |
Dataset | extended-cmu |
Submission date | Oct. 16, 2019, 8:07 a.m. |
Last modified | Nov. 4, 2019, 1:37 p.m. |
Publication url | https://ieeexplore.ieee.org/abstract/document/8463150 |
Code url | http://openaccess.thecvf.com/content_ICCV_2019/html/Larsson_Fine-Grained_Segmentation_Networks_Self-Supervised_Segmentation_for_Improved_Long-Term_Visual_Localization_ICCV_2019_paper.html |
Other info | FGSN 200 classes, long baseline, no reclustering, 1.5% of MVS map |
Detailed results for all conditions
Method | urban | suburban | park | overcast | sunny | foliage | mixed foliage | no foliage | low sun | cloudy | snow |
---|---|---|---|---|---|---|---|---|---|---|---|
PFSL FGSN 200 classes (sequential localization) | 94.1 / 99.3 / 100.0 | 100.0 / 100.0 / 100.0 | 97.6 / 99.9 / 99.9 | 96.4 / 99.7 / 100.0 | 96.3 / 99.8 / 100.0 | 96.8 / 99.8 / 100.0 | 97.5 / 99.6 / 100.0 | 98.6 / 99.9 / 99.9 | 97.8 / 99.6 / 100.0 | 99.0 / 100.0 / 100.0 | 98.5 / 99.9 / 99.9 |