- Task 1 - Track 1 – Single-domain training (Models must be trained exclusively on the published Cityscapes dataset) - Method: PixOOD
- Method info
- Samples list
- Per sample details
method: PixOOD2024-08-20
Authors: Tomáš Vojíř , Jan Šochman , and Jiří Matas
Affiliation: Czech Technical University in Prague
Description: This method is as reported in "PixOOD: Pixel-Level Out-of-Distribution Detection" paper that was accepted to ECCV 2024.
For details see the arXiv pre-print or the source code.