mlsanityData to 7 Oct 2026

CSFlow: Learning Optical Flow via Cross Strip Correlation for Autonomous Driving

arXiv 2202.00909 ↗First posted 2 Feb 2022cs.CV cs.RO eess.IVcited by 16

Not on the map. Why?

Abstract (arXiv)

Optical flow estimation is an essential task in self-driving systems, which helps autonomous vehicles perceive temporal continuity information of surrounding scenes. The calculation of all-pair correlation plays an important role in many existing state-of-the-art optical flow estimation methods. However, the reliance on local knowledge often limits the model's accuracy under complex street scenes. In this paper, we propose a new deep network architecture for optical flow estimation in autonomous driving--CSFlow, which consists of two novel modules: Cross Strip Correlation module (CSC) and Correlation Regression Initialization module (CRI). CSC utilizes a striping operation across the target image and the attended image to encode global context into correlation volumes, while maintaining high efficiency. CRI is used to maximally exploit the global context for optical flow initialization. Our method has achieved state-of-the-art accuracy on the public autonomous driving dataset KITTI-2015. Code is publicly available at https://github.com/MasterHow/CSFlow.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 29 of its 41 references that are in the map. First posted on arXiv 2 Feb 2022.

Builds on

29 of 41 references.

References · 29

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
  3. 2018CBAM: Convolutional Block Attention Modulecited by 2,463
  4. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  5. 2017Non-local Neural Networkscited by 2,290
  6. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
  7. 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 966
  8. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  9. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 780
  10. 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 673
  11. 2021RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matchingcited by 348
  12. 2016Deep Feature Flow for Video Recognitioncited by 186
17 more
  1. 2016Long Short-Term Memory-Networks for Machine Readingcited by 161
  2. 2019Iterative Residual Refinement for Joint Optical Flow and Occlusion Estimationcited by 131
  3. 2019Learning the Depths of Moving People by Watching Frozen Peoplecited by 120
  4. 2020MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Maskcited by 117
  5. 2018Models Matter, So Does Training: An Empirical Study of CNNs for Optical Flow Estimationcited by 94
  6. 2018Hierarchical Discrete Distribution Decomposition for Match Density Estimationcited by 93
  7. 2015EpicFlow: Edge-Preserving Interpolation of Correspondences for Optical Flowcited by 80
  8. 2019A Lightweight Optical Flow CNN - Revisiting Data Fidelity and Regularizationcited by 76
  9. 2017Semantic Video CNNs through Representation Warpingcited by 65
  10. 2021High-Resolution Optical Flow from 1D Attention and Correlationcited by 54
  11. 2020Displacement-Invariant Matching Cost Learning for Accurate Optical Flow Estimationcited by 49
  12. 2020LiteFlowNet3: Resolving Correspondence Ambiguity for More Accurate Optical Flow Estimationcited by 37
  13. 2020ScopeFlow: Dynamic Scene Scoping for Optical Flowcited by 24
  14. 2021VOLDOR: Visual Odometry from Log-logistic Dense Optical flow Residualscited by 17
  15. 2018Devon: Deformable Volume Network for Learning Optical Flowcited by 16
  16. 2016Can Ground Truth Label Propagation from Video help Semantic Segmentation?cited by 9
  17. 2020Learnable Cost Volume Using the Cayley Representationcited by 7

Built on by

Cited by year (2026 to 7 Oct)

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

16 papers.

Sort:

  1. 2023Multimodal Industrial Anomaly Detection via Hybrid Fusioncited by 138
  2. 2022Review on Panoramic Imaging and Its Applications in Scene Understandingcited by 42
  3. 2022Disentangling Architecture and Training for Optical Flowcited by 32
  4. 2022PanoFlow: Learning 360deg Optical Flow for Surrounding Temporal Understandingcited by 26
  5. 2023Towards Anytime Optical Flow Estimation with Event Camerascited by 15
  6. 2023StreamFlow: Streamlined Multi-Frame Optical Flow Estimation for Video Sequencescited by 12
  7. 2023On Pixel-level Performance Assessment in Anomaly Detectioncited by 6
  8. 2024Incomplete Multimodal Industrial Anomaly Detection via Cross-Modal Distillationcited by 6
  9. 2023FocusFlow: Boosting Key-Points Optical Flow Estimation for Autonomous Drivingcited by 5
  10. 2025PriOr-Flow: Enhancing Primitive Panoramic Optical Flow with Orthogonal Viewcited by 4
  11. 2024Benchmarking the Robustness of Optical Flow Estimation to Corruptionscited by 3
  12. 2024EI-Nexus: Towards Unmediated and Flexible Inter-Modality Local Feature Extraction and Matching for Event-Image Datacited by 1
4 more
  1. 2025An Expert Ensemble for Detecting Anomalous Scenes, Interactions, and Behaviors in Autonomous Drivingcited by 1
  2. 2025Moving Object Detection from Moving Camera Using Focus of Expansion Likelihood and Segmentationcited by 0
  3. 2026FLIGHT: Fibonacci Lattice-based Inference for Geometric Heading in real-Timecited by 0
  4. 2026Motion-guided sparse correction enables expert-quality point tracking across diverse microscopy regimescited by 0