Masked γ-SSL: learning uncertainty estimation via masked image modeling
This work proposes a semantic segmentation network that produces high-quality uncertainty estimates in a single forward pass. We exploit general representations from foundation models and unlabelled datasets through a Masked Image Modeling (MIM) approach, which is robust to augmentation hyper-parame...
Главные авторы: | Williams, DSW, Gadd, M, Newman, P, De Martini, D |
---|---|
Формат: | Conference item |
Язык: | English |
Опубликовано: |
IEEE
2024
|
Схожие документы
-
Mitigating distributional shift in semantic segmentation via uncertainty estimation from unlabeled data
по: Williams, DSW, и др.
Опубликовано: (2024) -
Fool me once: robust selective segmentation via out-of-distribution detection with contrastive learning
по: Williams, DSW, и др.
Опубликовано: (2021) -
Visual masking estimation based on structural uncertainty
по: Lin, Weisi, и др.
Опубликовано: (2013) -
To exploit uncertainty masking for adaptive image rendering
по: Dong, Lu, и др.
Опубликовано: (2013) -
Toward accurate hand mesh estimation via masked image modeling
по: Yanli Li, и др.
Опубликовано: (2025-01-01)