Searching for Contextual Subtasks for Semantic Segmentation
PúblicoConteúdo disponível para baixar
open in viewerDeep learning approaches for semantic segmentation have achieved tremendous success through their ability to model long-range scene context. However, by training for per-pixel classification, these methods fail to address the issue of class imbalance in segmentation datasets. Our work approaches segmentation from a contextual modeling standpoint by introducing novel subtasks as a human-provided hint or an auxiliary training signal. We first validate candidate subtasks through human-in-the-loop techniques to correct mistakes in segmentation, improving the mIoU of UPerNet-ResNet50 from 42.05 to 48.70 without any trained parameters. We then demonstrate the potential for multi-task learning of these subtasks with segmentation through a study of task gradients and end-to-end training.
- Creator
- Colaboradores
- Degree
- Unit
- Publisher
- Identifier
- etd-65241
- Palavra-chave
- Advisor
- Defense date
- Year
- 2022
- Date created
- 2022-04-28
- Resource type
- Rights statement
- Última modificação
- 2023-08-10
Relações
- Em Collection:
Itens
Itens
Miniatura | Título | Acesso | Embargo Release Date | Ações |
---|---|---|---|---|
cjmclaughlin_thesis_final.pdf | Público | Baixar |
Permanent link to this page: https://digital.wpi.edu/show/fb494c84s