A multimodal 3D foundation model pretrained on LSM volumes via masked reconstruction and image-text alignment enables improved few-shot segmentation, classification, and deblurring.
arXiv:2003.07311 (2022)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Multimodal 3D Foundation Model for Light Sheet Fluorescence Microscopy Enables Few-Shot Segmentation, Classification, and Deblurring
A multimodal 3D foundation model pretrained on LSM volumes via masked reconstruction and image-text alignment enables improved few-shot segmentation, classification, and deblurring.