TripNet: Learning Large-scale High-fidelity 3D Car Aerodynamics with Triplane Networks
This addresses the need for scalable and accurate surrogate models in engineering simulations, offering a more efficient alternative to traditional CFD solvers, though it appears incremental as it builds on implicit representation ideas.
The paper tackles the problem of accelerating Computational Fluid Dynamics (CFD) simulations for 3D car aerodynamics by introducing TripNet, a triplane-based neural framework that implicitly encodes geometry to avoid memory and resolution limitations of existing methods. It achieves state-of-the-art performance on datasets like DrivAerNet, accurately predicting drag coefficients, surface pressure, and full 3D flow fields.
Surrogate modeling has emerged as a powerful tool to accelerate Computational Fluid Dynamics (CFD) simulations. Existing 3D geometric learning models based on point clouds, voxels, meshes, or graphs depend on explicit geometric representations that are memory-intensive and resolution-limited. For large-scale simulations with millions of nodes and cells, existing models require aggressive downsampling due to their dependence on mesh resolution, resulting in degraded accuracy. We present TripNet, a triplane-based neural framework that implicitly encodes 3D geometry into a compact, continuous feature map with fixed dimension. Unlike mesh-dependent approaches, TripNet scales to high-resolution simulations without increasing memory cost, and enables CFD predictions at arbitrary spatial locations in a query-based fashion, independent of mesh connectivity or predefined nodes. TripNet achieves state-of-the-art performance on the DrivAerNet and DrivAerNet++ datasets, accurately predicting drag coefficients, surface pressure, and full 3D flow fields. With a unified triplane backbone supporting multiple simulation tasks, TripNet offers a scalable, accurate, and efficient alternative to traditional CFD solvers and existing surrogate models.