DisINR improves INR medical reconstruction by disentangling shared and subject-specific representations, pre-training the shared modules from raw data via differentiable forward models, and freezing them at test time.
Title resolution pending
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
-
Disentangled Learning Improves Implicit Neural Representations for Medical Reconstruction
DisINR improves INR medical reconstruction by disentangling shared and subject-specific representations, pre-training the shared modules from raw data via differentiable forward models, and freezing them at test time.