SF-VD generates labeled fluoroscopy videos via separate scene and motion diffusion models and reports improved guidewire segmentation, but test-set contamination weakens the evidence.
K.; Odonchimed, S.; and Xiao, F
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Label-Efficient Data Augmentation with Video Diffusion Models for Guidewire Segmentation in Cardiac Fluoroscopy
SF-VD generates labeled fluoroscopy videos via separate scene and motion diffusion models and reports improved guidewire segmentation, but test-set contamination weakens the evidence.