A new reliability score computed from the IoU difference between class-specific and class-agnostic heatmaps, boosted by adversarial enhancement, detects false negatives in binary industrial defect detectors with up to 100% recall.
Generalized out-of-distribution detection: A survey
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
A multitask learning method with instance-aware pseudo-labeling and class-focused contrastive learning outperforms prior UDA and WDA techniques for weakly supervised domain-adaptive mitochondria segmentation in EM images.
citing papers explorer
-
When Can We Trust Deep Neural Networks? Towards Reliable Industrial Deployment with an Interpretability Guide
A new reliability score computed from the IoU difference between class-specific and class-agnostic heatmaps, boosted by adversarial enhancement, detects false negatives in binary industrial defect detectors with up to 100% recall.
-
Instance-Aware Pseudo-Labeling and Class-Focused Contrastive Learning for Weakly Supervised Domain Adaptive Segmentation of Electron Microscopy
A multitask learning method with instance-aware pseudo-labeling and class-focused contrastive learning outperforms prior UDA and WDA techniques for weakly supervised domain-adaptive mitochondria segmentation in EM images.