mlsanityData to 14 Oct 2026

Unpaired Image Captioning via Scene Graph Alignments

arXiv 1903.10658 ↗First posted 26 Mar 2019cs.CVcited by 82

Not on the map. Why?

Abstract (arXiv)

Most of current image captioning models heavily rely on paired image-caption datasets. However, getting large scale image-caption paired data is labor-intensive and time-consuming. In this paper, we present a scene graph-based approach for unpaired image captioning. Our framework comprises an image scene graph generator, a sentence scene graph generator, a scene graph encoder, and a sentence decoder. Specifically, we first train the scene graph encoder and the sentence decoder on the text modality. To align the scene graphs between images and sentences, we propose an unsupervised feature alignment method that maps the scene graph features from the image to the sentence modality. Experimental results show that our proposed model can generate quite promising results without using any image-caption training pairs, outperforming existing methods by a wide margin.

Key facts. Cited by 82 papers in the map (data through 14 Oct 2026). First posted on arXiv 26 Mar 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 19; 2022: 25; 2023: 17; 2024: 11; 2025: 6; 2026: 4.

82 papers.

Sort:

By area:

  1. 2021From Show to Tell: A Survey on Deep Learning-based Image Captioningcited by 138
  2. 2021A Comprehensive Survey of Scene Graphs: Generation and Applicationcited by 125
  3. 2021Energy-Based Learning for Scene Graph Generationcited by 104
  4. 2022RelTR: Relation Transformer for Scene Graph Generationcited by 96
  5. 2022Stacked Hybrid-Attention and Group Collaborative Learning for Unbiased Scene Graph Generationcited by 67
  6. 2023Self-Supervised Multimodal Learning: A Surveycited by 58
  7. 2022Fine-Grained Scene Graph Generation with Data Transfercited by 57
  8. 2022VLMixer: Unpaired Vision-Language Pre-training via Cross-Modal CutMixcited by 45
  9. 2022Deep Learning Approaches on Image Captioning: A Reviewcited by 43
  10. 2023Incremental 3D Semantic Scene Graph Prediction from RGB Sequencescited by 39
  11. 2021In Defense of Scene Graphs for Image Captioningcited by 31
  12. 2023Expanding Scene Graph Boundaries: Fully Open-vocabulary Scene Graph Generation via Visual-Concept Alignment and Retentioncited by 29
38 more
  1. 2021Consensus Graph Representation Learning for Better Grounded Image Captioningcited by 28
  2. 2022RU-Net: Regularized Unrolling Network for Scene Graph Generationcited by 28
  3. 2021MAGIC: Multimodal relAtional Graph adversarIal inferenCe for Diverse and Unpaired Text-based Image Captioningcited by 27
  4. 2022Fine-Grained Predicates Learning for Scene Graph Generationcited by 27
  5. 2021Visual Distant Supervision for Scene Graph Generationcited by 25
  6. 2021Segmentation-grounded Scene Graph Generationcited by 25
  7. 2022Iterative Scene Graph Generationcited by 24
  8. 2021Exploiting Cross-Modal Prediction and Relation Consistency for Semi-Supervised Image Captioningcited by 23
  9. 2022Prompt-based Learning for Unpaired Image Captioningcited by 23
  10. 2023Unbiased Scene Graph Generation via Two-stage Causal Modelingcited by 21
  11. 2022NICEST: Noisy Label Correction and Training for Robust Scene Graph Generationcited by 19
  12. 2021Learning Structural Representations for Recipe Generation and Food Retrievalcited by 17
  13. 2023Text-based Person Search without Parallel Image-Text Datacited by 17
  14. 2023Information Screening whilst Exploiting! Multimodal Relation Extraction with Feature Denoising and Multimodal Topic Modelingcited by 16
  15. 2021Image Scene Graph Generation (SGG) Benchmarkcited by 15
  16. 2021Removing Word-Level Spurious Alignment between Images and Pseudo-Captions in Unsupervised Image Captioningcited by 14
  17. 2022Adaptive Fine-Grained Predicates Learning for Scene Graph Generationcited by 14
  18. 2022Label Semantic Knowledge Distillation for Unbiased Scene Graph Generationcited by 14
  19. 2022Aligning Source Visual and Target Language Domains for Unpaired Video Captioningcited by 14
  20. 2025A Causal Adjustment Module for Debiasing Scene Graph Generationcited by 14
  21. 2021Induce, Edit, Retrieve: Language Grounded Multimodal Schema for Instructional Video Retrievalcited by 13
  22. 2023MultiCapCLIP: Auto-Encoding Prompts for Zero-Shot Multilingual Visual Captioningcited by 13
  23. 2021Relational Graph Learning for Grounded Video Description Generationcited by 12
  24. 2021Predicate correlation learning for scene graph generationcited by 11
  25. 2023CgT-GAN: CLIP-guided Text GAN for Image Captioningcited by 10
  26. 2021Object-Centric Unsupervised Image Captioningcited by 9
  27. 2022Spatial Commonsense Graph for Object Localisation in Partial Scenescited by 9
  28. 2022Rethinking the Evaluation of Unbiased Scene Graph Generationcited by 9
  29. 2023Environment-Invariant Curriculum Relation Learning for Fine-Grained Scene Graph Generationcited by 9
  30. 2024MedRAT: Unpaired Medical Report Generation via Auxiliary Taskscited by 9
  31. 2022Unpaired Image Captioning by Image-level Weakly-Supervised Visual Concept Recognitioncited by 7
  32. 2022Swinv2-Imagen: Hierarchical Vision Transformer Diffusion Models for Text-to-Image Generationcited by 7
  33. 2024Unleashing Text-to-Image Diffusion Prior for Zero-Shot Image Captioningcited by 7
  34. 2021Exploring Semantic Relationships for Unpaired Image Captioningcited by 6
  35. 2024Towards Lifelong Scene Graph Generation with Knowledge-ware In-context Prompt Learningcited by 6
  36. 2021"Wikily" Supervised Neural Translation Tailored to Cross-Lingual Taskscited by 5
  37. 2022Paraphrasing Is All You Need for Novel Object Captioningcited by 4
  38. 2022CLID: Controlled-Length Image Descriptions with Limited Datacited by 4

50 most cited of 82 All 82 (JSON)