A two-phase self-training pipeline using classifier thresholds and DPO preferences adapts IntrinsicAnything to real-world albedo estimation.
Curriculum learning for vehicle lateral stability estimations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.GR 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
IntrinsicReal: Adapting IntrinsicAnything from Synthetic to Real Objects
A two-phase self-training pipeline using classifier thresholds and DPO preferences adapts IntrinsicAnything to real-world albedo estimation.