mlsanityData to 7 Oct 2026

Global Transport for Fluid Reconstruction with Learned Self-Supervision

arXiv 2104.06031 ↗First posted 13 Apr 2021cs.CV physics.flu-dyncited by 18

Not on the map. Why?

Abstract (arXiv)

We propose a novel method to reconstruct volumetric flows from sparse views via a global transport formulation. Instead of obtaining the space-time function of the observations, we reconstruct its motion based on a single initial state. In addition we introduce a learned self-supervision that constrains observations from unseen angles. These visual constraints are coupled via the transport constraints and a differentiable rendering step to arrive at a robust end-to-end reconstruction algorithm. This makes the reconstruction of highly realistic flow motions possible, even from only a single input view. We show with a variety of synthetic and real flows that the proposed global reconstruction of the transport process yields an improved reconstruction of the fluid motion.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 22 of its 82 references that are in the map. First posted on arXiv 13 Apr 2021.

Builds on

22 of 82 references.

References · 22

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  4. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  5. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  6. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  7. 2017Neural 3D Mesh Renderercited by 437
  8. 2018DeepVoxels: Learning Persistent 3D Feature Embeddingscited by 401
  9. 2018Learning Category-Specific Mesh Reconstruction from Image Collectionscited by 317
  10. 2020State of the Art on Neural Renderingcited by 276
  11. 2018The relativistic discriminator: a key element missing from standard GANcited by 234
  12. 2020Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-Solverscited by 201
10 more
  1. 2020How Much Position Information Do Convolutional Neural Networks Encode?cited by 171
  2. 2018Escaping Plato's Cave: 3D Shape From Adversarial Renderingcited by 170
  3. 2020Learning to Control PDEs with Differentiable Physicscited by 131
  4. 2017V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Mapcited by 102
  5. 2018SPNets: Differentiable Fluid Dynamics for Deep Neural Networkscited by 62
  6. 2019Sparse Surface Constraints for Combining Physics-based Elasticity Simulation and Correspondence-Free Object Reconstructioncited by 12
  7. 2020Lagrangian Neural Style Transfer for Fluidscited by 11
  8. 2019Transport-Based Neural Style Transfer for Smoke Simulationscited by 10
  9. 2018Coupled Fluid Density and Motion from Single Viewscited by 5
  10. 2017Single-image Tomography: 3D Volumes from 2D Cranial X-Rayscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 2; 2023: 5; 2024: 2; 2025: 5; 2026: 3.

18 papers.

Sort:

By area:

  1. 2022Physics Informed Neural Fields for Smoke Reconstruction with Sparse Datacited by 52
  2. 2022State of the Art in Dense Monocular Non-Rigid 3D Reconstructioncited by 30
  3. 2023Uncertainty-aware Surrogate Models for Airfoil Flow Simulations with Denoising Diffusion Probabilistic Modelscited by 22
  4. 2025FluidNexus: 3D Fluid Reconstruction and Prediction from a Single Videocited by 21
  5. 2023Inferring Hybrid Neural Fluid Fields from Videoscited by 18
  6. 2023Fluid Simulation on Neural Flow Mapscited by 12
  7. 2024Physics-Informed Learning of Characteristic Trajectories for Smoke Reconstructioncited by 9
  8. 2023Learning to Estimate Single-View Volumetric Flow Motions without 3D Supervisioncited by 5
  9. 2021Image Based Reconstruction of Liquids from 2D Surface Detectionscited by 4
  10. 2024Data-Efficient Inference of Neural Fluid Fields via SciML Foundation Modelcited by 4
  11. 2025Representing Flow Fields with Divergence-Free Kernels for Reconstructioncited by 4
  12. 2023Volumetric Cloud Field Reconstructioncited by 2
6 more
  1. 2025Learning an Implicit Physics Model for Image-based Fluid Simulationcited by 1
  2. 2026SurfPhase: 3D Interfacial Dynamics in Two-Phase Flows from Sparse Videoscited by 1
  3. 2026LagrangianSplats: Divergence-Free Transport of Gaussian Primitives for Fluid Reconstructioncited by 1
  4. 2025Light Transport-aware Diffusion Posterior Sampling for Single-View Reconstruction of 3D Volumescited by 0
  5. 2025SmokeSVD: Smoke Reconstruction from A Single View via Progressive Novel View Synthesis and Refinement with Diffusion Modelscited by 0
  6. 2026SplashSplat: Reconstructing Splashing Liquids from Real-World Multi-View Videoscited by 0