mlsanityData to 14 Oct 2026

WAFT: Warping-Alone Field Transforms for Optical Flow

arXiv 2506.21526 ↗First posted 26 Jun 2025cs.CVcited by 34

Not on the map. Why?

Abstract (arXiv)

We introduce Warping-Alone Field Transforms (WAFT), a simple and effective method for optical flow. WAFT is similar to RAFT but replaces cost volume with high-resolution warping, achieving better accuracy with lower memory cost. This design challenges the conventional wisdom that constructing cost volumes is necessary for strong performance. WAFT is a simple and flexible meta-architecture with minimal inductive biases and reliance on custom designs. Compared with existing methods, WAFT ranks 1st on Spring, Sintel, and KITTI benchmarks, achieves the best zero-shot generalization on KITTI, while being 1.3-4.1x faster than existing methods that have competitive accuracy (e.g., 1.3x than Flowformer++, 4.1x than CCMR+). Code and model weights are available at hrefhttps://github.com/princeton-vl/WAFThttps://github.com/princeton-vl/WAFT.

Key facts. Cited by 34 papers in the map (data through 14 Oct 2026). Builds on 49 of its 71 references that are in the map. First posted on arXiv 26 Jun 2025.

Builds on

49 of 71 references.

References · 49

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2023Segment Anythingcited by 8,234
  5. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  6. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  7. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  8. 2021Vision Transformers for Dense Predictioncited by 1,803
  9. 2024Depth Anything V2cited by 1,422
  10. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,059
  11. 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 968
  12. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
37 more
  1. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 784
  2. 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 506
  3. 2016Optical Flow Estimation using a Spatial Pyramid Networkcited by 424
  4. 2021GMFlow: Learning Optical Flow via Global Matchingcited by 327
  5. 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 275
  6. 2022Unifying Flow, Stereo and Depth Estimationcited by 275
  7. 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 224
  8. 2017Computer Vision for Autonomous Vehicles: Problems, Datasets and State of the Artcited by 182
  9. 2024SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flowcited by 174
  10. 2023Spring: A High-Resolution High-Detail Dataset and Benchmark for Scene Flow, Optical Flow and Stereocited by 172
  11. 2017Playing for Benchmarkscited by 169
  12. 2022CroCo v2: Improved Cross-view Completion Pre-training for Stereo Matching and Optical Flowcited by 166
  13. 2022CroCo: Self-Supervised Pre-training for 3D Vision Tasks by Cross-View Completioncited by 160
  14. 2019Quadratic video interpolationcited by 145
  15. 2020MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Maskcited by 117
  16. 2023The Surprising Effectiveness of Diffusion Models for Optical Flow and Monocular Depth Estimationcited by 116
  17. 2023FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimationcited by 114
  18. 2018Models Matter, So Does Training: An Empirical Study of CNNs for Optical Flow Estimationcited by 94
  19. 2017Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognitioncited by 82
  20. 2022CRAFT: Cross-Attentional Flow Transformer for Robust Optical Flowcited by 82
  21. 2017Accurate Optical Flow via Direct Cost Volume Processingcited by 66
  22. 2022Global Matching with Overlapping Attention for Optical Flow Estimationcited by 64
  23. 2022SKFlow: Learning Optical Flow with Super Kernelscited by 56
  24. 2018Representation Flow for Action Recognitioncited by 51
  25. 2022Multiview Stereo with Cascaded Epipolar RAFTcited by 51
  26. 2022DIP: Deep Inverse Patchmatch for High-Resolution Optical Flowcited by 40
  27. 2016Full Flow: Optical Flow Estimation By Global Optimization over Regular Gridscited by 34
  28. 2023Rethinking Optical Flow from Geometric Matching Consistent Perspectivecited by 33
  29. 2025DPFlow: Adaptive Optical Flow Estimation with a Dual-Pyramid Frameworkcited by 30
  30. 2022View Synthesis with Sculpted Neural Pointscited by 20
  31. 2023GAFlow: Incorporating Gaussian Attention into Optical Flowcited by 18
  32. 2023AnyFlow: Arbitrary Scale Optical Flow with Implicit Neural Representationcited by 17
  33. 2023SAMFlow: Eliminating Any Fragmentation in Optical Flow with Segment Anything Modelcited by 15
  34. 2023CCMR: High Resolution Optical Flow Estimation via Coarse-to-Fine Context-Guided Motion Reasoningcited by 15
  35. 2023Win-Win: Training High-Resolution Vision Transformers from Two Windowscited by 12
  36. 2023Memory-Efficient Optical Flow via Radius-Distribution Orthogonal Cost Volumecited by 9
  37. 2024Hybrid Cost Volume for Memory-Efficient Optical Flowcited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 33.

34 papers.

Sort:

By area:

  1. 2026CoWTracker: Tracking by Warping instead of Correlationcited by 14
  2. 2026MegaFlow: Zero-Shot Large Displacement Optical Flowcited by 5
  3. 2026WAFT-Stereo: Warping-Alone Field Transforms for Stereo Matchingcited by 3
  4. 2026SimpleProc: Fully Procedural Synthetic Data from Simple Rules for Multi-View Stereocited by 2
  5. 2026GeoFlow: Enforcing Implicit Geometric Consistency in Video Generationcited by 2
  6. 2026DynTrace: Tracking Dynamic Object Evidence for 4D Spatio-Temporal Reasoning in MLLMscited by 2
  7. 20264DGS-WAM: Bridging Past and Future with an Object-Centric World Action Model based on 4D Gaussian Splattingcited by 2
  8. 2025Towards Reliable Sea Ice Drift Estimation in the Arctic Deep Learning Optical Flow on RADARSAT-2cited by 1
  9. 2026SurfPhase: 3D Interfacial Dynamics in Two-Phase Flows from Sparse Videoscited by 1
  10. 2026PoseFM: Relative Camera Pose Estimation Through Flow Matchingcited by 1
  11. 2026TrAction: Action Recognition with Sparse Trajectoriescited by 1
  12. 2026ViSTR-Bench: Can MLLMs Reason from Continuous Visual Cues in Dynamic Scenes?cited by 1
22 more
  1. 2026MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligencecited by 1
  2. 2026Match4Annotate: Cross-Video Annotation Transfer in Ultrasound via Implicit Feature Flow-Guided Matchingcited by 0
  3. 2026RetimeGS: Continuous-Time Reconstruction of 4D Gaussian Splattingcited by 0
  4. 2026FlowIt: Global Matching via Hierarchical Transformers and Optimal Transport for Optical Flowcited by 0
  5. 2026DynaVid: Learning to Generate Highly Dynamic Videos using Synthetic Motion Datacited by 0
  6. 2026Flow of Truth: Proactive Temporal Forensics for Image-to-Video Generationcited by 0
  7. 2026JacobianAvatar: Temporally Consistent Semi-rigid Avatar Reconstruction from a Monocular Videocited by 0
  8. 2026On the Real-World Generalisability of Optical Flow Modelscited by 0
  9. 2026Point Tracking in Surgery--The 2025 Surgical Tattoos in Infrared Challenge (STIRC2025)cited by 0
  10. 2026WAVE-Stereo: Warp-Aligned Volume Encoding for Stereo Matchingcited by 0
  11. 2026Optical Flow from Photonscited by 0
  12. 2026Surg-UniWorld: A Unified Surgical World Model with Multimodal Control Expertscited by 0
  13. 2026WiFlow: Estimating Optical Flow using WiFi Channel State Informationcited by 0
  14. 2026FreeFlow: A Bias-free Hierarchical Transformer for Optical Flow Estimationcited by 0
  15. 2026TopoRig: Topology-Agnostic Facial Rigging via Multi-Source Supervisioncited by 0
  16. 2026Bi-FlowGS: Bridging Generative View Completion and Gaussian Geometry through Bidirectional Flow Co-Refinementcited by 0
  17. 2026TrackEverything: Long Horizon Dense Tracking via De-Duplicating 3D Scene Representationscited by 0
  18. 2026Unifying Video Tasks via Spatiotemporal Analogycited by 0
  19. 20263D Point Tracking with State Space Modelscited by 0
  20. 2026E-WAVE: Event-based Continuous Optical Flow via Warping-Aligned Visual Encodingcited by 0
  21. 2026DuSRFlow: Dual-Lens Super-Resolution via Parallax-Aware Flow Estimation and Complementary Reference Warpingcited by 0
  22. 2026Barycentric Guidance: Turning Foundation Image Editors into Continuous Affective Controllerscited by 0