mlsanityData to 14 Oct 2026

Improving Visual Recognition with Hyperbolical Visual Hierarchy Mapping

arXiv 2404.00974 ↗First posted 1 Apr 2024cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

Visual scenes are naturally organized in a hierarchy, where a coarse semantic is recursively comprised of several fine details. Exploring such a visual hierarchy is crucial to recognize the complex relations of visual elements, leading to a comprehensive scene understanding. In this paper, we propose a Visual Hierarchy Mapper (Hi-Mapper), a novel approach for enhancing the structured understanding of the pre-trained Deep Neural Networks (DNNs). Hi-Mapper investigates the hierarchical organization of the visual scene by 1) pre-defining a hierarchy tree through the encapsulation of probability densities; and 2) learning the hierarchical relations in hyperbolic space with a novel hierarchical contrastive loss. The pre-defined hierarchy tree recursively interacts with the visual features of the pre-trained DNNs through hierarchy decomposition and encoding procedures, thereby effectively identifying the visual hierarchy and enhancing the recognition of an entire scene. Extensive experiments demonstrate that Hi-Mapper significantly enhances the representation capability of DNNs, leading to an improved performance on various tasks, including image classification and dense prediction tasks.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 55 of its 70 references that are in the map. First posted on arXiv 1 Apr 2024.

Builds on

55 of 70 references.

References · 55

  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. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  5. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  6. 2017Mask R-CNNcited by 7,277
  7. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  8. 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
  9. 2021Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutionscited by 1,962
  10. 2018Unified Perceptual Parsing for Scene Understandingcited by 1,271
  11. 2021CvT: Introducing Convolutions to Vision Transformerscited by 947
  12. 2021CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classificationcited by 806
38 more
  1. 2021PVT v2: Improved Baselines with Pyramid Vision Transformercited by 751
  2. 2022Exploring Plain Vision Transformer Backbones for Object Detectioncited by 724
  3. 2017Poincar'e Embeddings for Learning Hierarchical Representationscited by 684
  4. 2019Panoptic Feature Pyramid Networkscited by 645
  5. 2019Stand-Alone Self-Attention in Vision Modelscited by 542
  6. 2021CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windowscited by 527
  7. 2021MViTv2: Improved Multiscale Vision Transformers for Classification and Detectioncited by 519
  8. 2015Variational Dropout and the Local Reparameterization Trickcited by 490
  9. 2019Hyperbolic Graph Convolutional Neural Networkscited by 402
  10. 2020Exploring Self-attention for Image Recognitioncited by 313
  11. 2021CMT: Convolutional Neural Networks Meet Vision Transformerscited by 310
  12. 2018Learning Continuous Hierarchies in the Lorentz Model of Hyperbolic Geometrycited by 265
  13. 2019Hyperbolic Image Embeddingscited by 221
  14. 2021Multi-Scale Vision Longformer: A New Vision Transformer for High-Resolution Image Encodingcited by 207
  15. 2017Detecting and Recognizing Human-Object Interactionscited by 199
  16. 2019Probabilistic Face Embeddingscited by 173
  17. 2018Poincar'e GloVe: Hyperbolic Word Embeddingscited by 156
  18. 2021Probabilistic Embeddings for Cross-Modal Retrievalcited by 151
  19. 2023Hyperbolic Image-Text Representationscited by 136
  20. 2022QuadTree Attention for Vision Transformerscited by 129
  21. 2022Hyperbolic Vision Transformers: Combining Improvements in Metric Learningcited by 111
  22. 2022Hyperbolic Image Segmentationcited by 102
  23. 2019SCOPS: Self-Supervised Co-Part Segmentationcited by 100
  24. 2014Word Representations via Gaussian Embeddingcited by 96
  25. 2021Affordance Transfer Learning for Human-Object Interaction Detectioncited by 79
  26. 2023Knowing Where to Focus: Event-aware Transformer for Video Groundingcited by 79
  27. 2021Unsupervised Part Discovery from Contrastive Reconstructioncited by 57
  28. 2022Pin the Memory: Learning to Generalize Semantic Segmentationcited by 49
  29. 2021Unsupervised Discovery of the Long-Tail in Instance Segmentation Using Hierarchical Self-Supervisioncited by 39
  30. 2020Hierarchical Human Parsing with Typed Part-Relation Reasoningcited by 38
  31. 2020Learning Compositional Neural Information Fusion for Human Parsingcited by 36
  32. 2021Probabilistic Modeling of Semantic Ambiguity for Scene Graph Generationcited by 35
  33. 2022Probabilistic Representations for Video Contrastive Learningcited by 34
  34. 2020HyperText: Endowing FastText with Hyperbolic Geometrycited by 28
  35. 2022Beyond Fixation: Dynamic Window Visual Transformercited by 28
  36. 2022HIER: Metric Learning Beyond Class Labels via Hierarchical Regularizationcited by 26
  37. 2023Probabilistic Prompt Learning for Dense Predictioncited by 17
  38. 2022Learning Hierarchical Image Segmentation For Recognition and By Recognitioncited by 16

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 4; 2026: 6.

11 papers.

  1. 2025Hyperbolic Dataset Distillationcited by 15
  2. 2024Adversarial Attacks on Hyperbolic Networkscited by 3
  3. 2026Fast and Geometrically Grounded Lorentz Neural Networkscited by 3
  4. 2025LIHE: Linguistic Instance-Split Hyperbolic-Euclidean Framework for Generalized Weakly-Supervised Referring Expression Comprehensioncited by 1
  5. 2025HypeVPR: Exploring Hyperbolic Space for Perspective to Equirectangular Visual Place Recognitioncited by 0
  6. 2025H3Former: Hypergraph-based Semantic-Aware Aggregation via Hyperbolic Hierarchical Contrastive Loss for Fine-Grained Visual Classificationcited by 0
  7. 2026Geometry-Aware State Space Model: A New Paradigm for Whole-Slide Image Representationcited by 0
  8. 2026Text-Driven Fusion for Infrared and Visible Images: Achieving Image Scene Adaptation on Hyperbolic Spacecited by 0
  9. 2026HSA: Hierarchical Slot Attention for Multi-granularity Scene-Decompositioncited by 0
  10. 2026HyperANFIS: Enhancing Rule Representation and Interpretability in Adaptive Neuro-Fuzzy Systems via Hyperbolic Geometrycited by 0
  11. 2026What Drives Hierarchy-Aware Image Retrieval? Taxonomy Alignment, Objective Choice, and Geometrycited by 0