FIGConv uses factorized 3D grids and global convolutions to predict car surface pressure and drag, reporting R2=0.957 on DrivAerNet drag and 0.89% pressure error on Ahmed body.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Factorized Implicit Global Convolution for Automotive Computational Fluid Dynamics Prediction
FIGConv uses factorized 3D grids and global convolutions to predict car surface pressure and drag, reporting R2=0.957 on DrivAerNet drag and 0.89% pressure error on Ahmed body.