mlsanityData to 7 Oct 2026

DynamicStereo: Consistent Dynamic Depth from Stereo Videos

arXiv 2305.02296 ↗First posted 3 May 2023cs.CV cs.AIcited by 147

Not on the map. Why?

Abstract (arXiv)

We consider the problem of reconstructing a dynamic scene observed from a stereo camera. Most existing methods for depth from stereo treat different stereo frames independently, leading to temporally inconsistent depth predictions. Temporal consistency is especially important for immersive AR or VR scenarios, where flickering greatly diminishes the user experience. We propose DynamicStereo, a novel transformer-based architecture to estimate disparity for stereo videos. The network learns to pool information from neighboring frames to improve the temporal consistency of its predictions. Our architecture is designed to process stereo videos efficiently through divided attention layers. We also introduce Dynamic Replica, a new benchmark dataset containing synthetic videos of people and animals in scanned environments, which provides complementary training and evaluation data for dynamic stereo closer to real applications than existing datasets. Training with this dataset further improves the quality of predictions of our proposed DynamicStereo as well as prior methods. Finally, it acts as a benchmark for consistent stereo methods.

Key facts. Cited by 147 papers in the map (data through 7 Oct 2026). Builds on 37 of its 54 references that are in the map. First posted on arXiv 3 May 2023.

Builds on

37 of 54 references.

References · 37

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  5. 2020End-to-End Object Detection with Transformerscited by 7,384
  6. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  7. 2021Vision Transformers for Dense Predictioncited by 1,789
  8. 2021Is Space-Time Attention All You Need for Video Understanding?cited by 1,609
  9. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,484
  10. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  11. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
  12. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
25 more
  1. 2019The Replica Dataset: A Digital Replica of Indoor Spacescited by 885
  2. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  3. 2016Playing for Data: Ground Truth from Computer Gamescited by 816
  4. 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 780
  5. 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 494
  6. 2018Pyramid Stereo Matching Networkcited by 486
  7. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 359
  8. 2021RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matchingcited by 348
  9. 2019GA-Net: Guided Aggregation Net for End-to-end Stereo Matchingcited by 242
  10. 2019Group-wise Correlation Stereo Networkcited by 235
  11. 2020Consistent Video Depth Estimationcited by 220
  12. 2022Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlationcited by 197
  13. 2020Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformerscited by 190
  14. 2020Robust Consistent Video Depth Estimationcited by 188
  15. 2017Playing for Benchmarkscited by 169
  16. 2015Delving Deeper into Convolutional Networks for Learning Video Representationscited by 146
  17. 2019Learning the Depths of Moving People by Watching Frozen Peoplecited by 120
  18. 2020HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matchingcited by 120
  19. 2019Domain-invariant Stereo Matching Networkscited by 108
  20. 2021AutoFlow: Learning a Better Training Set for Optical Flowcited by 102
  21. 2018Falling Things: A Synthetic Dataset for 3D Object Detection and Pose Estimationcited by 92
  22. 2014Computing the Stereo Matching Cost with a Convolutional Neural Networkcited by 74
  23. 2019Web Stereo Video Supervision for Depth Prediction from Dynamic Scenescited by 73
  24. 2020Bi3D: Stereo Depth Estimation via Binary Classificationscited by 34
  25. 2021Temporally Consistent Online Depth Estimation in Dynamic Scenescited by 26

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 14; 2025: 60; 2026: 72.

147 papers.

Sort:

By area:

  1. 2025Continuous 3D Perception Model with Persistent Statecited by 527
  2. 2023CoTracker: It is Better to Track Togethercited by 516
  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. 2024DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videoscited by 202
  6. 2025UniDepthV2: Universal Monocular Metric Depth Estimation Made Simplercited by 170
  7. 2025FoundationStereo: Zero-Shot Stereo Matchingcited by 168
  8. 2024CoTracker3: Simpler and Better Point Tracking by Pseudo-Labelling Real Videoscited by 122
  9. 2024Stereo4D: Learning How Things Move in 3D from Internet Stereo Videoscited by 89
  10. 2025Easi3R: Estimating Disentangled Motion from DUSt3R Without Trainingcited by 88
  11. 2025St4RTrack: Simultaneous 4D Reconstruction and Tracking in the Worldcited by 88
  12. 2025SpatialTrackerV2: 3D Point Tracking Made Easycited by 84
38 more
  1. 2025STream3R: Scalable Sequential 3D Reconstruction with Causal Transformercited by 76
  2. 2025TAPIP3D: Tracking Any Point in Persistent 3D Geometrycited by 67
  3. 2025UniK3D: Universal Camera Monocular 3D Estimationcited by 60
  4. 2025OmniWorld: A Multi-Domain and Multi-Modal Dataset for 4D World Modelingcited by 59
  5. 2024Depth Any Video with Scalable Synthetic Datacited by 57
  6. 2025AllTracker: Efficient Dense Point Tracking at High Resolutioncited by 54
  7. 2024Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videoscited by 48
  8. 2025Efficiently Reconstructing Dynamic Scenes One D4RT at a Timecited by 46
  9. 2024Stereo Anywhere: Robust Zero-Shot Deep Stereo Matching Even Where Either Stereo or Mono Failcited by 45
  10. 2026NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videoscited by 39
  11. 2025GeometryCrafter: Consistent Geometry Estimation for Open-world Videos with Diffusion Priorscited by 36
  12. 2025Segment Any Motion in Videoscited by 34
  13. 2025Any4D: Unified Feed-Forward Metric 4D Reconstructioncited by 32
  14. 2025MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Secondcited by 31
  15. 2025MVSAnywhere: Zero-Shot Multi-View Stereocited by 26
  16. 2024Match-Stereo-Videos: Bidirectional Alignment for Consistent Dynamic Stereo Matchingcited by 24
  17. 2025POMATO: Marrying Pointmap Matching with Temporal Motion for Dynamic 3D Reconstructioncited 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. 2025DA2: Depth Anything in Any Directioncited by 22
  22. 2025Zero-Shot Monocular Scene Flow Estimation in the Wildcited by 20
  23. 2026VGGT-Omegacited by 20
  24. 2025MonSter++: Unified Stereo Matching, Multi-view Stereo, and Real-time Stereo with Monodepth Priorscited by 19
  25. 2026InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fieldscited by 19
  26. 20264RC: 4D Reconstruction via Conditional Querying Anytime and Anywherecited by 17
  27. 2025FlashDepth: Real-time Streaming Video Depth Estimation at 2K Resolutioncited by 14
  28. 2026VGG-T3: Offline Feed-Forward 3D Reconstruction at Scalecited by 14
  29. 2025L4P: Towards Unified Low-Level 4D Vision Perceptioncited by 12
  30. 2024Match Stereo Videos via Bidirectional Alignmentcited by 11
  31. 2025JointDiT: Enhancing RGB-Depth Joint Modeling with Diffusion Transformerscited by 11
  32. 2025Can Video Diffusion Model Reconstruct 4D Geometry?cited by 10
  33. 2026Track4World: Feedforward World-centric Dense 3D Tracking of All Pixelscited by 10
  34. 2025Tracktention: Leveraging Point Tracking to Attend Videos Faster and Bettercited by 8
  35. 20254D-VGGT: A General Foundation Model with SpatioTemporal Awareness for Dynamic Scene Geometry Estimationcited by 8
  36. 2026Flow4R: Unifying 4D Reconstruction and Tracking with Scene Flowcited by 8
  37. 2026MoRe: Motion-aware Feed-forward 4D Reconstruction Transformercited by 8
  38. 2025Stereo Any Video: Temporally Consistent Stereo Matchingcited by 7

50 most cited of 147 All 147 (JSON)