# Factorized Implicit Global Convolution for Automotive Computational Fluid Dynamics Prediction

arXiv: https://arxiv.org/abs/2502.04317 · HTML: https://mlsanity.com/paper/2502.04317/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Feb 2025.

## Built on by (most cited first, 20 of 20)

- [GeoTransolver: Learning Physics on Irregular Domains Using Multi-scale Geometry Aware Physics Attention Transformer](https://mlsanity.com/paper/2512.20399/) (2025)
- [Geometric Operator Learning with Optimal Transport](https://mlsanity.com/paper/2507.20065/) (2025)
- [TripNet: Learning Large-scale High-fidelity 3D Car Aerodynamics with Triplane Networks](https://mlsanity.com/paper/2503.17400/) (2025)
- [CarBench: A Comprehensive Benchmark for Neural Surrogates on High-Fidelity 3D Car Aerodynamics](https://mlsanity.com/paper/2512.07847/) (2025)
- [DrivAerStar: An Industrial-Grade CFD Dataset for Vehicle Aerodynamic Optimization](https://mlsanity.com/paper/2510.16857/) (2025)
- [PhysGen: Physically Grounded 3D Shape Generation for Industrial Design](https://mlsanity.com/paper/2512.00422/) (2025)
- [A Benchmarking Framework for AI models in Automotive Aerodynamics](https://mlsanity.com/paper/2507.10747/) (2025)
- [GIST: Gauge-Invariant Spectral Transformers for Scalable Graph Neural Operators](https://mlsanity.com/paper/2603.16849/) (2026)
- [AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling](https://mlsanity.com/paper/2605.05586/) (2026)
- [3D Neural Operator-Based Flow Surrogates around 3D geometries: Signed Distance Functions and Derivative Constraints](https://mlsanity.com/paper/2503.17289/) (2025)
- [A Mixture of Experts Gating Network for Enhanced Surrogate Modeling in External Aerodynamics](https://mlsanity.com/paper/2508.21249/) (2025)
- [Car Drag Coefficient Prediction from 3D Point Clouds Using a Slice-Based Surrogate Model](https://mlsanity.com/paper/2601.02112/) (2026)
- [ShardTensor: Domain Parallelism for Scientific Machine Learning](https://mlsanity.com/paper/2605.11111/) (2026)
- [A Geometry-Aware Triplane Field Network for Vehicle Aerodynamic Prediction](https://mlsanity.com/paper/2606.07724/) (2026)
- [Learning Mappings in Mesh-based Simulations](https://mlsanity.com/paper/2506.12652/) (2025)
- [PMRT: A Training Recipe for Fast, 3D High-Resolution Aerodynamic Prediction](https://mlsanity.com/paper/2509.17182/) (2025)
- [CarCrashNet: A Large-Scale Dataset and Hierarchical Neural Solver for Data-Driven Structural Crash Simulation](https://mlsanity.com/paper/2605.07098/) (2026)
- [FSAN: Flow State Attention Network for Aerodynamic Prediction](https://mlsanity.com/paper/2609.06660/) (2026)
- [HGPTrans: Hierarchical Graph-Pooling Transolver for Automotive Aerodynamic Drag Coefficient Prediction](https://mlsanity.com/paper/2609.31765/) (2026)
- [CANTO: CAD-Native Transformer Operators for AI-Aided Engineering](https://mlsanity.com/paper/2609.36806/) (2026)

## Abstract (arXiv)

Computational Fluid Dynamics (CFD) is crucial for automotive design, requiring the analysis of large 3D point clouds to study how vehicle geometry affects pressure fields and drag forces. However, existing deep learning approaches for CFD struggle with the computational complexity of processing high-resolution 3D data. We propose Factorized Implicit Global Convolution (FIGConv), a novel architecture that efficiently solves CFD problems for very large 3D meshes with arbitrary input and output geometries. FIGConv achieves quadratic complexity O(N2), a significant improvement over existing 3D neural CFD models that require cubic complexity O(N3). Our approach combines Factorized Implicit Grids to approximate high-resolution domains, efficient global convolutions through 2D reparameterization, and a U-shaped architecture for effective information gathering and integration. We validate our approach on the industry-standard Ahmed body dataset and the large-scale DrivAerNet dataset. In DrivAerNet, our model achieves an R2 value of 0.95 for drag prediction, outperforming the previous state-of-the-art by a significant margin. This represents a 40% improvement in relative mean squared error and a 70% improvement in absolute mean squared error over previous methods.