.png){kind=icon}
Return
Segmentation Pseudolabel Generation Using the Multiple Instance Learning Choquet Integral
.png){kind=icon}
PRE
AI
DOI：10.1109/TFUZZ.2023.3338049.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Weakly supervised target detection and semantic segmentation (WSSS) approaches aim at learning object or pixel-level classification labels from imprecise, uncertain, or ambiguous data annotations. A crucial step in WSSS is to produce pseudolabels, which can be used to train a fully supervised semantic classifier. Post hoc attention mechanisms, such as class activation mapping (CAM), are commonly used to produce these pseudolabels. While traditional CAM methods derive feature importance from heuristics on gradient information, this work alternatively investigates whether a subset of discriminative activation feature maps can be down-selected and fused to improve pseudolabel accuracy. More specifically, the multiple instance Choquet integral (MICI) [1] was explored as a method for discriminative feature selection and fusion. Experimental results on both synthetic and real-world datasets indicate the utility of the MICI in down-selecting class-discriminative activation feature maps. Fusion of the MICI-selected sources was competitive to CAM methods for generating pseudolabels.
Keywords:
Attention
class activation map (CAM)
feature selection
fusion
multiple-instance learning (MIL)
pseudolabel
semantic segmentation
source selection
weakly supervised learning
Choquet integral (ChI)
Journal
IF:
11.9
Papers:
5.0K
Citations:
2.9W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSave
ShareSaveSynthesis of n-type semiconducting diamond film using diphosphorus pentaoxide as the doping source
PREAI
ShareSaveShareSave
View more