Box2Seg: Attention weighted loss and discriminative feature learning for weakly supervised segmentation
We propose a weakly supervised approach to semantic segmentation using bounding box annotations. Bounding boxes are treated as noisy labels for the foreground objects. We predict a per-class attention map that saliently guides the per-pixel cross entropy loss to focus on foreground pixels and refine...
Главные авторы: | Kulharia, V, Chandra, S, Agrawal, A, Torr, PHS, Tyagi, A |
---|---|
Формат: | Conference item |
Язык: | English |
Опубликовано: |
Springer International Publishing
2020
|
Схожие документы
-
Discovering class-specific pixels for weakly-supervised semantic segmentation
по: Chaudhry, A, и др.
Опубликовано: (2017) -
SegPGD: an effective and efficient adversarial attack for evaluating and boosting segmentation robustness
по: Gu, J, и др.
Опубликовано: (2022) -
Weakly- and semi-supervised panoptic segmentation
по: Li, Q, и др.
Опубликовано: (2018) -
IARS SegNet: Interpretable Attention Residual Skip Connection SegNet for Melanoma Segmentation
по: V. Shankara Narayanan, и др.
Опубликовано: (2024-01-01) -
Polarized Attention Weak Supervised Semantic Segmentation Network
по: Min Dai, и др.
Опубликовано: (2024-01-01)