mlsanityData to 14 Oct 2026

GraftNet: Towards Domain Generalized Stereo Matching with a Broad-Spectrum and Task-Oriented Feature

arXiv 2204.00179 ↗First posted 1 Apr 2022cs.CVcited by 43

Not on the map. Why?

Abstract (arXiv)

Although supervised deep stereo matching networks have made impressive achievements, the poor generalization ability caused by the domain gap prevents them from being applied to real-life scenarios. In this paper, we propose to leverage the feature of a model trained on large-scale datasets to deal with the domain shift since it has seen various styles of images. With the cosine similarity based cost volume as a bridge, the feature will be grafted to an ordinary cost aggregation module. Despite the broad-spectrum representation, such a low-level feature contains much general information which is not aimed at stereo matching. To recover more task-specific information, the grafted feature is further input into a shallow network to be transformed before calculating the cost. Extensive experiments show that the model generalization ability can be improved significantly with this broad-spectrum and task-oriented feature. Specifically, based on two well-known architectures PSMNet and GANet, our methods are superior to other robust algorithms when transferring from SceneFlow to KITTI 2015, KITTI 2012, and Middlebury. Code is available at https://github.com/SpadeLiu/Graft-PSMNet.

Key facts. Cited by 43 papers in the map (data through 14 Oct 2026). Builds on 32 of its 47 references that are in the map. First posted on arXiv 1 Apr 2022.

Builds on

32 of 47 references.

References · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,241
  5. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  6. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
  7. 2020Dense Contrastive Learning for Self-Supervised Visual Pre-Trainingcited by 512
  8. 2018Pyramid Stereo Matching Networkcited by 489
  9. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 360
  10. 2020Virtual KITTI 2cited by 353
  11. 2021RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matchingcited by 353
  12. 2019GA-Net: Guided Aggregation Net for End-to-end Stereo Matchingcited by 244
20 more
  1. 2019Group-wise Correlation Stereo Networkcited by 235
  2. 2020AANet: Adaptive Aggregation Network for Efficient Stereo Matchingcited by 202
  3. 2020Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformerscited by 190
  4. 2020Hierarchical Neural Architecture Search for Deep Stereo Matchingcited by 166
  5. 2021CFNet: Cascade and Fused Cost Volume for Robust Stereo Matchingcited by 154
  6. 2019Domain-invariant Stereo Matching Networkscited by 109
  7. 2018SegStereo: Exploiting Semantic Information for Disparity Estimationcited by 88
  8. 2018Real-time self-adaptive deep stereocited by 80
  9. 2014Computing the Stereo Matching Cost with a Convolutional Neural Networkcited by 74
  10. 2020Knowledge Distillation for Multi-task Learningcited by 51
  11. 2020Learning Stereo from Single Imagescited by 45
  12. 2019Learning to Adapt for Stereocited by 42
  13. 2019EdgeStereo: An Effective Multi-Task Learning Network for Stereo Matching and Edge Detectioncited by 39
  14. 2020Matching-space Stereo Networks for Cross-domain Generalizationcited by 32
  15. 2020AdaStereo: A Simple and Efficient Approach for Adaptive Stereo Matchingcited by 31
  16. 2019Guided Stereo Matchingcited by 24
  17. 2020StereoGAN: Bridging Synthetic-to-Real Domain Gap by Joint Optimization of Domain Translation and Stereo Matchingcited by 21
  18. 2018Practical Deep Stereo (PDS): Toward applications-friendly deep stereo matchingcited by 20
  19. 2021Neural Disparity Refinement for Arbitrary Resolution Stereocited by 20
  20. 2019Learning Across Tasks and Domainscited by 16

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 10; 2024: 14; 2025: 14; 2026: 5.

43 papers.

Sort:

By area:

  1. 2023CAT-Seg: Cost Aggregation for Open-Vocabulary Semantic Segmentationcited by 219
  2. 2023Iterative Geometry Encoding Volume for Stereo Matchingcited by 190
  3. 2025FoundationStereo: Zero-Shot Stereo Matchingcited by 173
  4. 2024IGEV++: Iterative Multi-range Geometry Encoding Volumes for Stereo Matchingcited by 59
  5. 2024Stereo Anywhere: Robust Zero-Shot Deep Stereo Matching Even Where Either Stereo or Mono Failcited by 47
  6. 2023NeRF-Supervised Deep Stereocited by 46
  7. 2023Booster: a Benchmark for Depth from Images of Specular and Transparent Surfacescited by 33
  8. 2025Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matchingcited by 28
  9. 2024A Survey on Deep Stereo Matching in the Twentiescited by 23
  10. 2023Adaptive Multi-Modal Cross-Entropy Loss for Stereo Matchingcited by 19
  11. 2023Revisiting Depth Completion from a Stereo Matching Perspective for Cross-domain Generalizationcited by 18
  12. 2023CGI-Stereo: Accurate and Real-Time Stereo Matching via Context and Geometry Interactioncited by 17
31 more
  1. 2024Unifying Feature and Cost Aggregation with Transformers for Semantic and Visual Correspondencecited by 14
  2. 2025Learning Robust Stereo Matching in the Wild with Selective Mixture-of-Expertscited by 14
  3. 2024Federated Online Adaptation for Deep Stereocited by 12
  4. 2025Diagnosing and Addressing Pitfalls in KG-RAG Datasets: Toward More Reliable Benchmarkingcited by 12
  5. 2024Learning Intra-view and Cross-view Geometric Knowledge for Stereo Matchingcited by 11
  6. 2025Diving into the Fusion of Monocular Priors for Generalized Stereo Matchingcited by 10
  7. 2023Active Stereo Without Pattern Projectorcited by 8
  8. 2024UWStereo: A Large Synthetic Dataset for Underwater Stereo Matchingcited by 7
  9. 2024Stereo Risk: A Continuous Modeling Approach to Stereo Matchingcited by 6
  10. 2026PromptStereo: Zero-Shot Stereo Matching via Structure and Motion Promptscited by 6
  11. 2024These Maps Are Made by Propagation: Adapting Deep Stereo Networks to Road Scenarios with Decisive Disparity Diffusioncited by 5
  12. 2024Stereo Anything: Unifying Zero-shot Stereo Matching with Large-Scale Mixed Datacited by 5
  13. 2023StereoFlowGAN: Co-training for Stereo and Flow with Unsupervised Domain Adaptationcited by 4
  14. 2024DCVSMNet: Double Cost Volume Stereo Matching Networkcited by 3
  15. 2025OV-COAST: Cost Aggregation with Optimal Transport for Open-Vocabulary Semantic Segmentationcited by 2
  16. 2025PanMatch: Unleashing the Potential of Large Vision Models for Unified Matching Modelscited by 2
  17. 2024SR-Stereo & DAPE: Stepwise Regression and Pre-trained Edges for Practical Stereo Matchingcited by 1
  18. 2023Towards Generalizable Referring Image Segmentation via Target Prompt and Visual Coherencecited by 0
  19. 2024SyntStereo2Real: Edge-Aware GAN for Remote Sensing Image-to-Image Translation while Maintaining Stereo Constraintcited by 0
  20. 2024Active Stereo in the Wild through Virtual Pattern Projectioncited by 0
  21. 2025MIDAS: Modeling Ground-Truth Distributions with Dark Knowledge for Domain Generalized Stereo Matchingcited by 0
  22. 2025CMD: Constraining Multimodal Distribution for Domain Adaptation in Stereo Matchingcited by 0
  23. 2025ESMStereo: Enhanced ShuffleMixer Disparity Upsampling for Real-Time and Accurate Stereo Matchingcited by 0
  24. 2025Preserving Guidance in Cost-Volume Retrieval under Extreme LiDAR Sparsity in Iterative Stereocited by 0
  25. 2025Domain Generalized Stereo Matching with Uncertainty-guided Data Augmentationcited by 0
  26. 2025MatchAttention: Embedding Explicit Matching Constraints into Attention for Efficient Stereo Matchingcited by 0
  27. 2025Generalized Geometry Encoding Volume for Real-time Stereo Matchingcited by 0
  28. 2026HyperWalker: Dynamic Hypergraph-Based Deep Diagnosis for Multi-Hop Clinical Modeling across EHR and X-Ray in Medical VLMscited by 0
  29. 2026Geometry Reinforced Efficient Attention Tuning Equipped with Normals for Robust Stereo Matchingcited by 0
  30. 2026Bidirectional Cross-Modal Prompting for Event-Frame Asymmetric Stereocited by 0
  31. 2026Acoustically Grounded Cost Learning for Open-Vocabulary Audio-Visual Semantic Segmentationcited by 0