# fVDB: A Deep-Learning Framework for Sparse, Large-Scale, and High-Performance Spatial Intelligence

arXiv: https://arxiv.org/abs/2407.01781 · HTML: https://mlsanity.com/paper/2407.01781/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 36 references that are in the map.
- First posted on arXiv 1 Jul 2024.

## Builds on (most cited first, 15 of 15)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields](https://mlsanity.com/paper/2111.12077/) (2021)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks](https://mlsanity.com/paper/1904.08755/) (2019)
- [PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection](https://mlsanity.com/paper/1912.13192/) (2019)
- [BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation](https://mlsanity.com/paper/2205.13542/) (2022)
- [Accelerating 3D Deep Learning with PyTorch3D](https://mlsanity.com/paper/2007.08501/) (2020)
- [XCube: Large-Scale 3D Generative Modeling using Sparse Voxel Hierarchies](https://mlsanity.com/paper/2312.03806/) (2023)
- [TorchSparse: Efficient Point Cloud Inference Engine](https://mlsanity.com/paper/2204.10319/) (2022)
- [TorchSparse++: Efficient Training and Inference Framework for Sparse Convolution on GPUs](https://mlsanity.com/paper/2311.12862/) (2023)

## Built on by (most cited first, 23 of 23)

- [Native and Compact Structured Latents for 3D Generation](https://mlsanity.com/paper/2512.14692/) (2025)
- [Dora: Sampling and Benchmarking for 3D Shape Variational Auto-Encoders](https://mlsanity.com/paper/2412.17808/) (2024)
- [SCube: Instant Large-Scale Scene Reconstruction using VoxSplats](https://mlsanity.com/paper/2410.20030/) (2024)
- [InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models](https://mlsanity.com/paper/2412.03934/) (2024)
- [STORM: Spatio-Temporal Reconstruction Model for Large-Scale Outdoor Scenes](https://mlsanity.com/paper/2501.00602/) (2024)
- [Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos](https://mlsanity.com/paper/2412.03526/) (2024)
- [Lyra 2.0: Explorable Generative 3D Worlds](https://mlsanity.com/paper/2604.13036/) (2026)
- [Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering](https://mlsanity.com/paper/2412.04459/) (2024)
- [ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Grasping](https://mlsanity.com/paper/2504.10857/) (2025)
- [EarthCrafter: Scalable 3D Earth Generation via Dual-Sparse Latent Diffusion](https://mlsanity.com/paper/2507.16535/) (2025)
- [ShapeShifter: 3D Variations Using Multiscale and Sparse Point-Voxel Diffusion](https://mlsanity.com/paper/2502.02187/) (2025)
- [Benchmarking Convolutional Neural Network and Graph Neural Network based Surrogate Models on a Real-World Car External Aerodynamics Dataset](https://mlsanity.com/paper/2504.06699/) (2025)
- [Gabor Fields: Orientation-Selective Level-of-Detail for Volume Rendering](https://mlsanity.com/paper/2602.05081/) (2026)
- [DynamicTree: Interactive Real Tree Animation via Sparse Voxel Spectrum](https://mlsanity.com/paper/2510.22213/) (2025)
- [Seen2Scene: Completing Realistic 3D Scenes with Visibility-Guided Flow](https://mlsanity.com/paper/2603.28548/) (2026)
- [GraspFoM: Towards Reconstruction-Driven Robotic Grasping with 3D Foundation Priors](https://mlsanity.com/paper/2606.08440/) (2026)
- [ArchComplete: Autoregressive 3D Architectural Design Generation with Hierarchical Diffusion-Based Upsampling](https://mlsanity.com/paper/2412.17957/) (2024)
- [PMRT: A Training Recipe for Fast, 3D High-Resolution Aerodynamic Prediction](https://mlsanity.com/paper/2509.17182/) (2025)
- [PointCNN++: Performant Convolution on Native Points](https://mlsanity.com/paper/2511.23227/) (2025)
- [SLIM-VDB: A Real-Time 3D Probabilistic Semantic Mapping Framework](https://mlsanity.com/paper/2512.12945/) (2025)
- [XGrid-Mapping: Explicit Implicit Hybrid Grid Submaps for Efficient Incremental Neural LiDAR Mapping](https://mlsanity.com/paper/2512.20976/) (2025)
- [AnyScene: Towards Highly Controllable Driving Scene Generation at Anywhere and Beyond](https://mlsanity.com/paper/2605.26113/) (2026)
- [A Scalable PyTorch Abstraction for Multi-GPU Gaussian Splatting](https://mlsanity.com/paper/2606.11390/) (2026)

## Abstract (arXiv)

We present fVDB, a novel GPU-optimized framework for deep learning on large-scale 3D data. fVDB provides a complete set of differentiable primitives to build deep learning architectures for common tasks in 3D learning such as convolution, pooling, attention, ray-tracing, meshing, etc. fVDB simultaneously provides a much larger feature set (primitives and operators) than established frameworks with no loss in efficiency: our operators match or exceed the performance of other frameworks with narrower scope. Furthermore, fVDB can process datasets with much larger footprint and spatial resolution than prior works, while providing a competitive memory footprint on small inputs. To achieve this combination of versatility and performance, fVDB relies on a single novel VDB index grid acceleration structure paired with several key innovations including GPU accelerated sparse grid construction, convolution using tensorcores, fast ray tracing kernels using a Hierarchical Digital Differential Analyzer algorithm (HDDA), and jagged tensors. Our framework is fully integrated with PyTorch enabling interoperability with existing pipelines, and we demonstrate its effectiveness on a number of representative tasks such as large-scale point-cloud segmentation, high resolution 3D generative modeling, unbounded scale Neural Radiance Fields, and large-scale point cloud reconstruction.