A dual-branch network with autocorrelation-based consistency prior for surface normal estimation from polarization in ice, plus a new real-world dataset, reports 16.01 deg MAE outperforming prior methods by 2.74 deg.
ISBN 978-3-031-72854-9 978-3-031-72855-6
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
-
Structure-Aware Consistency Priors for Shape from Polarization in Complex Media
A dual-branch network with autocorrelation-based consistency prior for surface normal estimation from polarization in ice, plus a new real-world dataset, reports 16.01 deg MAE outperforming prior methods by 2.74 deg.