mlsanityData to 7 Oct 2026

Spring: A High-Resolution High-Detail Dataset and Benchmark for Scene Flow, Optical Flow and Stereo

arXiv 2303.01943 ↗First posted 3 Mar 2023cs.CVcited by 170

Not on the map. Why?

Abstract (arXiv)

While recent methods for motion and stereo estimation recover an unprecedented amount of details, such highly detailed structures are neither adequately reflected in the data of existing benchmarks nor their evaluation methodology. Hence, we introduce Spring - a large, high-resolution, high-detail, computer-generated benchmark for scene flow, optical flow, and stereo. Based on rendered scenes from the open-source Blender movie "Spring", it provides photo-realistic HD datasets with state-of-the-art visual effects and ground truth training data. Furthermore, we provide a website to upload, analyze and compare results. Using a novel evaluation methodology based on a super-resolved UHD ground truth, our Spring benchmark can assess the quality of fine structures and provides further detailed performance statistics on different image regions. Regarding the number of ground truth frames, Spring is 60× larger than the only scene flow benchmark, KITTI 2015, and 15× larger than the well-established MPI Sintel optical flow benchmark. Initial results for recent methods on our benchmark show that estimating fine details is indeed challenging, as their accuracy leaves significant room for improvement. The Spring benchmark and the corresponding datasets are available at http://spring-benchmark.org.

Key facts. Cited by 170 papers in the map (data through 7 Oct 2026). Builds on 30 of its 52 references that are in the map. First posted on arXiv 3 Mar 2023.

Builds on

30 of 52 references.

References · 30

  1. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  2. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
  3. 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 966
  4. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  5. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 780
  6. 2016Optical Flow Estimation using a Spatial Pyramid Networkcited by 421
  7. 2022Kubric: A scalable dataset generatorcited by 373
  8. 2021RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matchingcited by 348
  9. 2020Virtual KITTI 2cited by 345
  10. 2016Virtual Worlds as Proxy for Multi-Object Tracking Analysiscited by 340
  11. 2021GMFlow: Learning Optical Flow via Global Matchingcited by 324
  12. 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 273
18 more
  1. 2019GA-Net: Guided Aggregation Net for End-to-end Stereo Matchingcited by 242
  2. 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 223
  3. 2022Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlationcited by 197
  4. 2017Playing for Benchmarkscited by 169
  5. 2020Hierarchical Neural Architecture Search for Deep Stereo Matchingcited by 166
  6. 2022Attention Concatenation Volume for Accurate and Efficient Stereo Matchingcited by 130
  7. 2020RAFT-3D: Scene Flow using Rigid-Motion Embeddingscited by 121
  8. 2021AutoFlow: Learning a Better Training Set for Optical Flowcited by 102
  9. 2021CamLiFlow: Bidirectional Camera-LiDAR Fusion for Joint Optical Flow and Scene Flow Estimationcited by 56
  10. 2018Learning Rigidity in Dynamic Scenes with a Moving Camera for 3D Motion Field Estimationcited by 38
  11. 2019Learning Multi-Human Optical Flowcited by 26
  12. 2017Optical Flow in Mostly Rigid Scenescited by 20
  13. 2018Motion-based Object Segmentation based on Dense RGB-D Scene Flowcited by 18
  14. 2022A Perturbation-Constrained Adversarial Attack for Evaluating the Robustness of Optical Flowcited by 16
  15. 2022M-FUSE: Multi-frame Fusion for Scene Flow Estimationcited by 14
  16. 2022High Resolution Multi-Scale RAFT (Robust Vision Challenge 2022)cited by 12
  17. 2022Semi-Supervised Learning of Optical Flow by Flow Supervisorcited by 9
  18. 2022Multi-Scale RAFT: Combining Hierarchical Concepts for Learning-based Optical FLow Estimationcited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 10; 2024: 25; 2025: 59; 2026: 76.

170 papers.

Sort:

By area:

  1. 2025Depth Anything 3: Recovering the Visual Space from Any Viewscited by 626
  2. 2024MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motioncited by 431
  3. 2024Depth Pro: Sharp Monocular Metric Depth in Less Than a Secondcited by 395
  4. 2025MapAnything: Universal Feed-Forward Metric 3D Reconstructioncited by 295
  5. 2024MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervisioncited by 290
  6. 2025MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Detailscited by 250
  7. 2024SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flowcited by 169
  8. 2025FoundationStereo: Zero-Shot Stereo Matchingcited by 168
  9. 2024Tora: Trajectory-oriented Diffusion Transformer for Video Generationcited by 161
  10. 2025Streaming 4D Visual Geometry Transformercited by 147
  11. 2023Infinite Photorealistic Worlds using Procedural Generationcited by 142
  12. 2025SpatialTrackerV2: 3D Point Tracking Made Easycited by 84
38 more
  1. 2025Point3R: Streaming 3D Reconstruction with Explicit Spatial Pointer Memorycited by 82
  2. 2024Align3R: Aligned Monocular Depth Estimation for Dynamic Videoscited by 79
  3. 2025Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstructioncited by 66
  4. 2024MemFlow: Optical Flow Estimation and Prediction with Memorycited by 55
  5. 2025AllTracker: Efficient Dense Point Tracking at High Resolutioncited by 54
  6. 2024Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videoscited by 48
  7. 2025Dens3R: A Foundation Model for 3D Geometry Predictioncited by 44
  8. 2026NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videoscited by 39
  9. 2025GeometryCrafter: Consistent Geometry Estimation for Open-world Videos with Diffusion Priorscited by 36
  10. 2025UFM: A Simple Path towards Unified Dense Correspondence with Flowcited by 35
  11. 2026LoGeR: Long-Context Geometric Reconstruction with Hybrid Memorycited by 33
  12. 2025WAFT: Warping-Alone Field Transforms for Optical Flowcited by 32
  13. 2025Any4D: Unified Feed-Forward Metric 4D Reconstructioncited by 32
  14. 2025MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Secondcited by 31
  15. 2025DPFlow: Adaptive Optical Flow Estimation with a Dual-Pyramid Frameworkcited by 30
  16. 2025Dynamic Camera Poses and Where to Find Themcited by 26
  17. 2026Masked Depth Modeling for Spatial Perceptioncited by 24
  18. 2024A Survey on Deep Stereo Matching in the Twentiescited by 23
  19. 2025Advances in Feed-Forward 3D Reconstruction and View Synthesis: A Surveycited by 23
  20. 2026ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Trainingcited by 23
  21. 2024SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillationcited by 21
  22. 2025Zero-Shot Monocular Scene Flow Estimation in the Wildcited by 20
  23. 2026LongStream: Long-Sequence Streaming Autoregressive Visual Geometrycited by 17
  24. 2023Distracting Downpour: Adversarial Weather Attacks for Motion Estimationcited by 15
  25. 2023SAMFlow: Eliminating Any Fragmentation in Optical Flow with Segment Anything Modelcited by 15
  26. 2025FlashDepth: Real-time Streaming Video Depth Estimation at 2K Resolutioncited by 14
  27. 2026VGG-T3: Offline Feed-Forward 3D Reconstruction at Scalecited by 14
  28. 2024BlinkVision: A Benchmark for Optical Flow, Scene Flow and Point Tracking Estimation using RGB Frames and Eventscited by 13
  29. 2023Win-Win: Training High-Resolution Vision Transformers from Two Windowscited by 12
  30. 2023StreamFlow: Streamlined Multi-Frame Optical Flow Estimation for Video Sequencescited by 12
  31. 2025L4P: Towards Unified Low-Level 4D Vision Perceptioncited by 12
  32. 2026CoWTracker: Tracking by Warping instead of Correlationcited by 12
  33. 2023Understanding Physical Dynamics with Counterfactual World Modelingcited by 11
  34. 2024LayeredFlow: A Real-World Benchmark for Non-Lambertian Multi-Layer Optical Flowcited by 11
  35. 2024VE-Bench: Subjective-Aligned Benchmark Suite for Text-Driven Video Editing Quality Assessmentcited by 10
  36. 2025D2USt3R: Enhancing 3D Reconstruction for Dynamic Scenescited by 10
  37. 2025FlowSeek: Optical Flow Made Easier with Depth Foundation Models and Motion Basescited by 10
  38. 2026MetricAnything: Scaling Metric Depth Pretraining with Noisy Heterogeneous Sourcescited by 10

50 most cited of 170 All 170 (JSON)