mlsanityData to 7 Oct 2026

CostFormer:Cost Transformer for Cost Aggregation in Multi-view Stereo

arXiv 2305.10320 ↗First posted 17 May 2023cs.CVcited by 9

Not on the map. Why?

Abstract (arXiv)

The core of Multi-view Stereo(MVS) is the matching process among reference and source pixels. Cost aggregation plays a significant role in this process, while previous methods focus on handling it via CNNs. This may inherit the natural limitation of CNNs that fail to discriminate repetitive or incorrect matches due to limited local receptive fields. To handle the issue, we aim to involve Transformer into cost aggregation. However, another problem may occur due to the quadratically growing computational complexity caused by Transformer, resulting in memory overflow and inference latency. In this paper, we overcome these limits with an efficient Transformer-based cost aggregation network, namely CostFormer. The Residual Depth-Aware Cost Transformer(RDACT) is proposed to aggregate long-range features on cost volume via self-attention mechanisms along the depth and spatial dimensions. Furthermore, Residual Regression Transformer(RRT) is proposed to enhance spatial attention. The proposed method is a universal plug-in to improve learning-based MVS methods.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 29 of its 44 references that are in the map. First posted on arXiv 17 May 2023.

Builds on

29 of 44 references.

References · 29

  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. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  5. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  6. 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 780
  7. 2019BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networkscited by 487
  8. 2019Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matchingcited by 414
  9. 2019Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inferencecited by 252
  10. 2020PatchmatchNet: Learned Multi-View Patchmatch Stereocited by 204
  11. 2020Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformerscited by 190
  12. 2019Cost Volume Pyramid Based Depth Inference for Multi-View Stereocited by 154
17 more
  1. 2019Deep Stereo using Adaptive Thin Volume Representation with Uncertainty Awarenesscited by 151
  2. 2019Point-Based Multi-View Stereo Networkcited by 147
  3. 2017SurfaceNet: An End-to-end 3D Neural Network for Multiview Stereopsiscited by 128
  4. 2021TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformerscited by 117
  5. 2020Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinementcited by 105
  6. 2020Visibility-aware Multi-view Stereo Networkcited by 97
  7. 2022Rethinking Depth Estimation for Multi-View Stereo: A Unified Representationcited by 82
  8. 2020Dense Hybrid Recurrent Multi-view Stereo Net with Dynamic Consistency Checkingcited by 81
  9. 2021CATs: Cost Aggregation Transformers for Visual Correspondencecited by 77
  10. 2021AA-RMVSNet: Adaptive Aggregation Recurrent Multi-view Stereo Networkcited by 71
  11. 2021IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereocited by 56
  12. 2022MVSTER: Epipolar Transformer for Efficient Multi-View Stereocited by 50
  13. 2019Pyramid Multi-view Stereo Net with Self-adaptive View Aggregationcited by 44
  14. 2020PVSNet: Pixelwise Visibility-Aware Multi-View Stereo Networkcited by 35
  15. 2022WT-MVSNet: Window-based Transformers for Multi-view Stereocited by 20
  16. 2021PatchMatch-RL: Deep MVS with Pixelwise Depth, Normal, and Visibilitycited by 17
  17. 2021Multi-View Stereo with Transformercited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 4; 2026: 3.

9 papers.

  1. 2024MVSFormer++: Revealing the Devil in Transformer's Details for Multi-View Stereocited by 49
  2. 2025Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentationcited by 4
  3. 2026PCA-Seg: Revisiting Cost Aggregation for Open-Vocabulary Semantic and Part Segmentationcited by 4
  4. 2024SDL-MVS: View Space and Depth Deformable Learning Paradigm for Multi-View Stereo Reconstruction in Remote Sensingcited by 3
  5. 2025MVSMamba: Multi-View Stereo with State Space Modelcited by 1
  6. 2025Detail-aware multi-view stereo network for depth estimationcited by 0
  7. 2025IA-MVS: Instance-Focused Adaptive Depth Sampling for Multi-View Stereocited by 0
  8. 2026Resolving Long-Tail Ambiguity in Unsupervised 3D Point Cloud Segmentation with Language Priorscited by 0
  9. 2026Spatio-Temporal Similarity Volume Aggregation for Open-Vocabulary Action Recognitioncited by 0