mlsanityData to 13 Oct 2026

Conformer: Local Features Coupling Global Representations for Visual Recognition

arXiv 2105.03889 ↗First posted 9 May 2021cs.CVcited by 164

Not on the map. Why?

Abstract (arXiv)

Within Convolutional Neural Network (CNN), the convolution operations are good at extracting local features but experience difficulty to capture global representations. Within visual transformer, the cascaded self-attention modules can capture long-distance feature dependencies but unfortunately deteriorate local feature details. In this paper, we propose a hybrid network structure, termed Conformer, to take advantage of convolutional operations and self-attention mechanisms for enhanced representation learning. Conformer roots in the Feature Coupling Unit (FCU), which fuses local features and global representations under different resolutions in an interactive fashion. Conformer adopts a concurrent structure so that local features and global representations are retained to the maximum extent. Experiments show that Conformer, under the comparable parameter complexity, outperforms the visual transformer (DeiT-B) by 2.3% on ImageNet. On MSCOCO, it outperforms ResNet-101 by 3.7% and 3.6% mAPs for object detection and instance segmentation, respectively, demonstrating the great potential to be a general backbone network. Code is available at https://github.com/pengzhiliang/Conformer.

Key facts. Cited by 164 papers in the map (data through 13 Oct 2026). Builds on 47 of its 60 references that are in the map. First posted on arXiv 9 May 2021.

Builds on

47 of 60 references.

References · 47

  1. 2017Attention Is All You Needcited by 60,106
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,950
  3. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,304
  4. 2020Language Models are Few-Shot Learnerscited by 32,320
  5. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,043
  6. 2014Microsoft COCO: Common Objects in Contextcited by 17,550
  7. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,321
  8. 2020End-to-End Object Detection with Transformerscited by 7,396
  9. 2017Mask R-CNNcited by 7,266
  10. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,991
  11. 2016Densely Connected Convolutional Networkscited by 6,707
  12. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,426
35 more
  1. 2016Layer Normalizationcited by 5,624
  2. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,170
  3. 2016Feature Pyramid Networks for Object Detectioncited by 5,027
  4. 2017Squeeze-and-Excitation Networkscited by 4,772
  5. 2020Training data-efficient image transformers & distillation through attentioncited by 4,574
  6. 2014Going Deeper with Convolutionscited by 4,239
  7. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,417
  8. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,882
  9. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,544
  10. 2018CBAM: Convolutional Block Attention Modulecited by 2,469
  11. 2015Learning Deep Features for Discriminative Localizationcited by 2,434
  12. 2017Non-local Neural Networkscited by 2,291
  13. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,266
  14. 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,512
  15. 2017Deformable Convolutional Networkscited by 1,486
  16. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,418
  17. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,383
  18. 2020Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformerscited by 1,369
  19. 2017Random Erasing Data Augmentationcited by 1,238
  20. 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,129
  21. 2021Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNetcited by 1,101
  22. 2016Deep Networks with Stochastic Depthcited by 977
  23. 2020Designing Network Design Spacescited by 956
  24. 2020Pre-Trained Image Processing Transformercited by 892
  25. 2020Quantifying Attention Flow in Transformerscited by 767
  26. 2021Bottleneck Transformers for Visual Recognitioncited by 364
  27. 2019Attention Augmented Convolutional Networkscited by 363
  28. 2020Visual Transformers: Token-based Image Representation and Processing for Computer Visioncited by 334
  29. 2019GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyondcited by 331
  30. 2017Relation Networks for Object Detectioncited by 305
  31. 2017Dilated Residual Networkscited by 294
  32. 2018Gather-Excite: Exploiting Feature Context in Convolutional Neural Networkscited by 138
  33. 2021TS-CAM: Token Semantic Coupled Attention Map for Weakly Supervised Object Localizationcited by 123
  34. 2020Toward Transformer-Based Object Detectioncited by 84
  35. 2020Rethinking Localization Map: Towards Accurate Object Perception with Self-Enhancement Mapscited by 15

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2021: 16; 2022: 42; 2023: 43; 2024: 37; 2025: 17; 2026: 9.

164 papers.

Sort:

By area:

  1. 2022Vision Transformer Adapter for Dense Predictionscited by 565
  2. 2021Mobile-Former: Bridging MobileNet and Transformercited by 234
  3. 2021ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Biascited by 216
  4. 2021HRFormer: High-Resolution Transformer for Dense Predictioncited by 149
  5. 2023Dual Aggregation Transformer for Image Super-Resolutioncited by 124
  6. 2022ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition and Beyondcited by 117
  7. 2023CompletionFormer: Depth Completion with Convolutions and Vision Transformerscited by 92
  8. 2021On the Integration of Self-Attention and Convolutioncited by 74
  9. 2022DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spottingcited by 74
  10. 2023A Comprehensive Survey on Applications of Transformers for Deep Learning Taskscited by 71
  11. 2023Radar-Camera Fusion for Object Detection and Semantic Segmentation in Autonomous Driving: A Comprehensive Reviewcited by 57
  12. 2022A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localizationcited by 56
38 more
  1. 2023IEBins: Iterative Elastic Bins for Monocular Depth Estimationcited by 53
  2. 2022Deepfake Detection: A Comprehensive Survey from the Reliability Perspectivecited by 49
  3. 2022Patch-Fool: Are Vision Transformers Always Robust Against Adversarial Perturbations?cited by 48
  4. 2023Neural Degradation Representation Learning for All-In-One Image Restorationcited by 46
  5. 2022MixFormer: Mixing Features across Windows and Dimensionscited by 43
  6. 2021Refiner: Refining Self-attention for Vision Transformerscited by 38
  7. 2023Xformer: Hybrid X-Shaped Transformer for Image Denoisingcited by 37
  8. 2023HRDFuse: Monocular 360degDepth Estimation by Collaboratively Learning Holistic-with-Regional Depth Distributionscited by 37
  9. 2021Pale Transformer: A General Vision Transformer Backbone with Pale-Shaped Attentioncited by 36
  10. 2023DilateFormer: Multi-Scale Dilated Transformer for Visual Recognitioncited by 34
  11. 2021Dual-stream Network for Visual Recognitioncited by 33
  12. 2021Expression Snippet Transformer for Robust Video-based Facial Expression Recognitioncited by 33
  13. 2022Transforming medical imaging with Transformers? A comparative review of key properties, current progresses, and future perspectivescited by 33
  14. 2022Locality Guidance for Improving Vision Transformers on Tiny Datasetscited by 33
  15. 2022HiFuse: Hierarchical Multi-Scale Feature Fusion Network for Medical Image Classificationcited by 30
  16. 2022Bridging the Gap Between Vision Transformers and Convolutional Neural Networks on Small Datasetscited by 28
  17. 2022TransCAM: Transformer Attention-based CAM Refinement for Weakly Supervised Semantic Segmentationcited by 26
  18. 2023Lightweight Vision Transformer with Bidirectional Interactioncited by 25
  19. 2021Semi-Supervised Vision Transformerscited by 24
  20. 2022Fast-iTPN: Integrally Pre-Trained Transformer Pyramid Network with Token Migrationcited by 24
  21. 2023Pedestrian Attribute Recognition via CLIP based Prompt Vision-Language Fusioncited by 24
  22. 2024ScribFormer: Transformer Makes CNN Work Better for Scribble-based Medical Image Segmentationcited by 24
  23. 2022HiViT: Hierarchical Vision Transformer Meets Masked Image Modelingcited by 23
  24. 2023WeakTr: Exploring Plain Vision Transformer for Weakly-supervised Semantic Segmentationcited by 22
  25. 2023Deeply-Coupled Convolution-Transformer with Spatial-temporal Complementary Learning for Video-based Person Re-identificationcited by 21
  26. 2022Cross-Modality High-Frequency Transformer for MR Image Super-Resolutioncited by 20
  27. 2022Boosting vision transformers for image retrievalcited by 19
  28. 2023Robustifying Token Attention for Vision Transformerscited by 19
  29. 2022SATr: Slice Attention with Transformer for Universal Lesion Detectioncited by 17
  30. 2022Few-Shot Learning Meets Transformer: Unified Query-Support Transformers for Few-Shot Classificationcited by 16
  31. 2023URCDC-Depth: Uncertainty Rectified Cross-Distillation with CutFlip for Monocular Depth Estimationcited by 16
  32. 2024Efficient Modulation for Vision Networkscited by 16
  33. 2023Improving Scene Text Image Super-resolution via Dual Prior Modulation Networkcited by 15
  34. 2023Transformers in Small Object Detection: A Benchmark and Survey of State-of-the-Artcited by 15
  35. 2024RingMo-Aerial: An Aerial Remote Sensing Foundation Model With Affine Transformation Contrastive Learningcited by 15
  36. 2021CSformer: Bridging Convolution and Transformer for Compressive Sensingcited by 14
  37. 2023Computation-efficient Deep Learning for Computer Vision: A Surveycited by 14
  38. 2022Centralized Feature Pyramid for Object Detectioncited by 13

50 most cited of 164 All 164 (JSON)