mlsanityData to 14 Oct 2026

Crucial-Diff: A Unified Diffusion Model for Crucial Image and Annotation Synthesis in Data-scarce Scenarios

arXiv 2507.09915 ↗First posted 14 Jul 2025cs.CVcited by 0

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

The scarcity of data in various scenarios, such as medical, industry and autonomous driving, leads to model overfitting and dataset imbalance, thus hindering effective detection and segmentation performance. Existing studies employ the generative models to synthesize more training samples to mitigate data scarcity. However, these synthetic samples are repetitive or simplistic and fail to provide "crucial information" that targets the downstream model's weaknesses. Additionally, these methods typically require separate training for different objects, leading to computational inefficiencies. To address these issues, we propose Crucial-Diff, a domain-agnostic framework designed to synthesize crucial samples. Our method integrates two key modules. The Scene Agnostic Feature Extractor (SAFE) utilizes a unified feature extractor to capture target information. The Weakness Aware Sample Miner (WASM) generates hard-to-detect samples using feedback from the detection results of downstream model, which is then fused with the output of SAFE module. Together, our Crucial-Diff framework generates diverse, high-quality training data, achieving a pixel-level AP of 83.63% and an F1-MAX of 78.12% on MVTec. On polyp dataset, Crucial-Diff reaches an mIoU of 81.64% and an mDice of 87.69%. Code is publicly available at https://github.com/JJessicaYao/Crucial-diff.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 52 of its 78 references that are in the map. First posted on arXiv 14 Jul 2025.

Builds on

52 of 78 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024InstaGen: Enhancing Object Detection by Training on Synthetic Dataset8 Feb
  2. 2023DatasetDM: Synthesizing Data with Perception Annotations Using Diffusion Models11 Aug
  3. 2022Multi-Concept Customization of Text-to-Image Diffusion8 Dec
  4. 2022Hierarchical Text-Conditional Image Generation with CLIP Latents13 Apr

Other references · 48

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2021Zero-Shot Text-to-Image Generationcited by 4,288
  5. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  6. 2016Improved Techniques for Training GANscited by 2,970
  7. 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,940
  8. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,245
  9. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 776
  10. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 694
  11. 2017AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networkscited by 664
  12. 2019Kvasir-SEG: A Segmented Polyp Datasetcited by 589
36 more
  1. 2016StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networkscited by 577
  2. 2021CutPaste: Self-Supervised Learning for Anomaly Detection and Localizationcited by 531
  3. 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 514
  4. 2020PraNet: Parallel Reverse Attention Network for Polyp Segmentationcited by 460
  5. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 421
  6. 2023ELITE: Encoding Visual Concepts into Textual Embeddings for Customized Text-to-Image Generationcited by 416
  7. 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 271
  8. 2022A Unified Model for Multi-class Anomaly Detectioncited by 268
  9. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  10. 2016A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Imagescited by 221
  11. 2021Few-shot Image Generation via Cross-domain Correspondencecited by 207
  12. 2021Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformerscited by 159
  13. 2024Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detectioncited by 139
  14. 2019Meta-Sim: Learning to Generate Synthetic Datasetscited by 128
  15. 2019Self-Ensembling with GAN-based Data Augmentation for Domain Adaptation in Semantic Segmentationcited by 126
  16. 2023AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Modelcited by 111
  17. 2022Catching Both Gray and Black Swans: Open-set Supervised Anomaly Detectioncited by 104
  18. 2024PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detectioncited by 98
  19. 2022Omni-frequency Channel-selection Representations for Unsupervised Anomaly Detectioncited by 80
  20. 2021Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspectioncited by 77
  21. 2022Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detectioncited by 72
  22. 2021Explainable Deep Few-shot Anomaly Detection with Deviation Networkscited by 69
  23. 2023Few-Shot Defect Image Generation via Defect-Aware Feature Manipulationcited by 67
  24. 2022Prototypical Residual Networks for Anomaly Detection and Localizationcited by 56
  25. 2024FiLo: Zero-Shot Anomaly Detection by Fine-Grained Description and High-Quality Localizationcited by 51
  26. 2023Towards Artistic Image Aesthetics Assessment: a Large-scale Dataset and a New Methodcited by 49
  27. 2023Exploring Plain ViT Reconstruction for Multi-class Unsupervised Anomaly Detectioncited by 47
  28. 2021SinGAN-Seg: Synthetic training data generation for medical image segmentationcited by 43
  29. 2023Lesion-aware Dynamic Kernel for Polyp Segmentationcited by 39
  30. 2020Night-time Scene Parsing with a Large Real Datasetcited by 37
  31. 2023ArSDM: Colonoscopy Images Synthesis with Adaptive Refinement Semantic Diffusion Modelscited by 31
  32. 2025Noise-Consistent Siamese-Diffusion for Medical Image Synthesis and Segmentationcited by 23
  33. 2023Mask-conditioned latent diffusion for generating gastrointestinal polyp imagescited by 16
  34. 2024iSeg: An Iterative Refinement-based Framework for Training-free Segmentationcited by 10
  35. 2024Learn From Zoom: Decoupled Supervised Contrastive Learning For WCE Image Classificationcited by 6
  36. 2023Single-Image-Based Deep Learning for Segmentation of Early Esophageal Cancer Lesionscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Context-guided Responsible Data Augmentation with Diffusion Models
  2. 2024Diff-Plugin: Revitalizing Details for Diffusion-based Low-level Tasks
  3. 2024Distribution-Aware Data Expansion with Diffusion Models
  4. 2023DiffuGen: Adaptable Approach for Generating Labeled Image Datasets using Stable Diffusion Models
  5. 2024High-Precision Dichotomous Image Segmentation via Probing Diffusion Capacity
  6. 2023DiffusionEngine: Diffusion Model is Scalable Data Engine for Object Detection
  7. 2025DiffDoctor: Diagnosing Image Diffusion Models Before Treating
  8. 2023DiffFit: Unlocking Transferability of Large Diffusion Models via Simple Parameter-Efficient Fine-Tuning