mlsanityData to 14 Oct 2026

L2G: A Simple Local-to-Global Knowledge Transfer Framework for Weakly Supervised Semantic Segmentation

arXiv 2204.03206 ↗First posted 7 Apr 2022cs.CVcited by 57

Not on the map. Why?

Abstract (arXiv)

Mining precise class-aware attention maps, a.k.a, class activation maps, is essential for weakly supervised semantic segmentation. In this paper, we present L2G, a simple online local-to-global knowledge transfer framework for high-quality object attention mining. We observe that classification models can discover object regions with more details when replacing the input image with its local patches. Taking this into account, we first leverage a local classification network to extract attentions from multiple local patches randomly cropped from the input image. Then, we utilize a global network to learn complementary attention knowledge across multiple local attention maps online. Our framework conducts the global network to learn the captured rich object detail knowledge from a global view and thereby produces high-quality attention maps that can be directly used as pseudo annotations for semantic segmentation networks. Experiments show that our method attains 72.1% and 44.2% mIoU scores on the validation set of PASCAL VOC 2012 and MS COCO 2014, respectively, setting new state-of-the-art records. Code is available at https://github.com/PengtaoJiang/L2G.

Key facts. Cited by 57 papers in the map (data through 14 Oct 2026). Builds on 50 of its 66 references that are in the map. First posted on arXiv 7 Apr 2022.

Builds on

50 of 66 references.

References · 50

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  5. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  6. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,241
  7. 2016Pyramid Scene Parsing Networkcited by 2,479
  8. 2015Learning Deep Features for Discriminative Localizationcited by 2,436
  9. 2021Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNetcited by 1,101
  10. 2014Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFscited by 554
  11. 2018Born Again Neural Networkscited by 546
  12. 2016RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentationcited by 465
38 more
  1. 2016ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentationcited by 331
  2. 2018Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentationcited by 321
  3. 2020Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentationcited by 300
  4. 2015What's the Point: Semantic Segmentation with Point Supervisioncited by 298
  5. 2019Structured Knowledge Distillation for Dense Predictioncited by 279
  6. 2019Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relationscited by 247
  7. 2015BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentationcited by 238
  8. 2019A Simple Pooling-Based Design for Real-Time Salient Object Detectioncited by 212
  9. 2017Object Region Mining with Adversarial Erasing: A Simple Classification to Semantic Segmentation Approachcited by 201
  10. 2016Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentationcited by 198
  11. 2015Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentationcited by 189
  12. 2016Wider or Deeper: Revisiting the ResNet Model for Visual Recognitioncited by 184
  13. 2018Tell Me Where to Look: Guided Attention Inference Networkcited by 169
  14. 2020Single-Stage Semantic Segmentation from Image Labelscited by 167
  15. 2018Adversarial Complementary Learning for Weakly Supervised Object Localizationcited by 166
  16. 2019FickleNet: Weakly and Semi-supervised Semantic Image Segmentation using Stochastic Inferencecited by 162
  17. 2018Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentationcited by 145
  18. 2020Weakly-Supervised Semantic Segmentation via Sub-category Explorationcited by 145
  19. 2020Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentationcited by 144
  20. 2021Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentationcited by 134
  21. 2021Railroad is not a Train: Saliency as Pseudo-pixel Supervision for Weakly Supervised Semantic Segmentationcited by 127
  22. 2019Knowledge Adaptation for Efficient Semantic Segmentationcited by 105
  23. 2018Learning random-walk label propagation for weakly-supervised semantic segmentationcited by 101
  24. 2018Self-Erasing Network for Integral Object Attentioncited by 95
  25. 2021Non-Salient Region Object Mining for Weakly Supervised Semantic Segmentationcited by 95
  26. 2019Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approachcited by 88
  27. 2018Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Featurescited by 81
  28. 2021Leveraging Auxiliary Tasks with Affinity Learning for Weakly Supervised Semantic Segmentationcited by 77
  29. 2021Context Decoupling Augmentation for Weakly Supervised Semantic Segmentationcited by 69
  30. 2019Self-Supervised Difference Detection for Weakly-Supervised Semantic Segmentationcited by 58
  31. 2020Group-Wise Semantic Mining for Weakly Supervised Semantic Segmentationcited by 52
  32. 2021Discriminative Region Suppression for Weakly-Supervised Semantic Segmentationcited by 50
  33. 2021Pseudo-mask Matters in Weakly-supervised Semantic Segmentationcited by 48
  34. 2014Fully Convolutional Multi-Class Multiple Instance Learningcited by 37
  35. 2019Saliency Guided Self-attention Network for Weakly and Semi-supervised Semantic Segmentationcited by 28
  36. 2020Mixup-CAM: Weakly-supervised Semantic Segmentation via Uncertainty Regularizationcited by 24
  37. 2016Bottom-Up Top-Down Cues for Weakly-Supervised Semantic Segmentationcited by 5
  38. 2020Multi-Miner: Object-Adaptive Region Mining for Weakly-Supervised Semantic Segmentationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 7; 2023: 25; 2024: 19; 2025: 4; 2026: 2.

57 papers.

Sort:

By area:

  1. 2022CLIP is Also an Efficient Segmenter: A Text-Driven Approach for Weakly Supervised Semantic Segmentationcited by 118
  2. 2023Token Contrast for Weakly-Supervised Semantic Segmentationcited by 65
  3. 2023Segment Anything Model (SAM) Enhanced Pseudo Labels for Weakly Supervised Semantic Segmentationcited by 53
  4. 2022A Survey on Label-efficient Deep Image Segmentation: Bridging the Gap between Weak Supervision and Dense Predictioncited by 46
  5. 2023CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentationcited by 44
  6. 2023Self-Calibrated Cross Attention Network for Few-Shot Segmentationcited by 32
  7. 2023Extracting Class Activation Maps from Non-Discriminative Features as wellcited by 30
  8. 2023Segment Anything is A Good Pseudo-label Generator for Weakly Supervised Semantic Segmentationcited by 27
  9. 2022Out-of-Candidate Rectification for Weakly Supervised Semantic Segmentationcited by 26
  10. 2023WeakTr: Exploring Plain Vision Transformer for Weakly-supervised Semantic Segmentationcited by 22
  11. 2023MCTformer+: Multi-Class Token Transformer for Weakly Supervised Semantic Segmentationcited by 22
  12. 2023An Alternative to WSSS? An Empirical Study of the Segment Anything Model (SAM) on Weakly-Supervised Semantic Segmentation Problemscited by 19
38 more
  1. 2023Weakly-Supervised Semantic Segmentation with Image-Level Labels: from Traditional Models to Foundation Modelscited by 19
  2. 2023Background Activation Suppression for Weakly Supervised Object Localization and Semantic Segmentationcited by 17
  3. 2024Spatial Structure Constraints for Weakly Supervised Semantic Segmentationcited by 17
  4. 2024DuPL: Dual Student with Trustworthy Progressive Learning for Robust Weakly Supervised Semantic Segmentationcited by 17
  5. 2023Multi-Granularity Denoising and Bidirectional Alignment for Weakly Supervised Semantic Segmentationcited by 16
  6. 2023Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Modelscited by 15
  7. 2024Frozen CLIP: A Strong Backbone for Weakly Supervised Semantic Segmentationcited by 15
  8. 2023USAGE: A Unified Seed Area Generation Paradigm for Weakly Supervised Semantic Segmentationcited by 14
  9. 2024Separate and Conquer: Decoupling Co-occurrence via Decomposition and Representation for Weakly Supervised Semantic Segmentationcited by 14
  10. 2023MARS: Model-agnostic Biased Object Removal without Additional Supervision for Weakly-Supervised Semantic Segmentationcited by 13
  11. 2024Knowledge Transfer with Simulated Inter-Image Erasing for Weakly Supervised Semantic Segmentationcited by 8
  12. 2025Exploring CLIP's Dense Knowledge for Weakly Supervised Semantic Segmentationcited by 8
  13. 2023Centered Self-Attention Layerscited by 7
  14. 2022Foreground-Background Separation through Concept Distillation from Generative Image Foundation Modelscited by 5
  15. 2024Disentangling spatio-temporal knowledge for weakly supervised object detection and segmentation in surgical videocited by 5
  16. 2022Dual Progressive Transformations for Weakly Supervised Semantic Segmentationcited by 4
  17. 2024Tackling Ambiguity from Perspective of Uncertainty Inference and Affinity Diversification for Weakly Supervised Semantic Segmentationcited by 4
  18. 2024Fine-grained Background Representation for Weakly Supervised Semantic Segmentationcited by 4
  19. 2024DIAL: Dense Image-text ALignment for Weakly Supervised Semantic Segmentationcited by 4
  20. 2023A Language-Guided Benchmark for Weakly Supervised Open Vocabulary Semantic Segmentationcited by 3
  21. 2023Prompting classes: Exploring the Power of Prompt Class Learning in Weakly Supervised Semantic Segmentationcited by 3
  22. 2023All-pairs Consistency Learning for Weakly Supervised Semantic Segmentationcited by 3
  23. 2024Semantic Prompt Learning for Weakly-Supervised Semantic Segmentationcited by 3
  24. 2024Weakly Supervised Co-training with Swapping Assignments for Semantic Segmentationcited by 3
  25. 2024DHR: Dual Features-Driven Hierarchical Rebalancing in Inter- and Intra-Class Regions for Weakly-Supervised Semantic Segmentationcited by 3
  26. 2023Exploit CAM by itself: Complementary Learning System for Weakly Supervised Semantic Segmentationcited by 2
  27. 2023Masked Collaborative Contrast for Weakly Supervised Semantic Segmentationcited by 2
  28. 2023Weakly Supervised Semantic Segmentation by Knowledge Graph Inferencecited by 2
  29. 2024Weakly-supervised Semantic Segmentation via Dual-stream Contrastive Learning of Cross-image Contextual Informationcited by 2
  30. 2024Pixel-Level Domain Adaptation: A New Perspective for Enhancing Weakly Supervised Semantic Segmentationcited by 2
  31. 2022RecurSeed and EdgePredictMix: Pseudo-Label Refinement Learning for Weakly Supervised Semantic Segmentation across Single- and Multi-Stage Frameworkscited by 1
  32. 2022ExpNet: A unified network for Expert-Level Classificationcited by 1
  33. 2025Mitigating Spurious Correlations in Weakly Supervised Semantic Segmentation via Cross-architecture Consistency Regularizationcited by 1
  34. 2025SynthSeg-Agents: Multi-Agent Synthetic Data Generation for Zero-Shot Weakly Supervised Semantic Segmentationcited by 1
  35. 2023Exploring Weakly Supervised Semantic Segmentation Ensembles for Medical Imaging Systemscited by 0
  36. 2023A Self-Training Framework Based on Multi-Scale Attention Fusion for Weakly Supervised Semantic Segmentationcited by 0
  37. 2023P-NOC: adversarial training of CAM generating networks for robust weakly supervised semantic segmentation priorscited by 0
  38. 2024P2Seg: Pointly-supervised Segmentation via Mutual Distillationcited by 0

50 most cited of 57 All 57 (JSON)