Few-Shot Synthetic Data Generation with Diffusion Models for Downstream Vision Tasks

• URL: http://arxiv.org/abs/2605.11898v1
• Date: Tue, 12 May 2026 10:11:57 GMT
• Title: Few-Shot Synthetic Data Generation with Diffusion Models for Downstream Vision Tasks
• Abstract summary: Class imbalance is a persistent challenge in visual recognition.<n>We propose a lightweight synthetic data augmentation pipeline that fine-tunes a LoRA adapter on as few as 20-50 real images of a rare class.
• Score: 36.94429692322632
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Class imbalance is a persistent challenge in visual recognition, particularly in safety-critical domains where collecting positive examples is expensive and rare events are inherently underrepresented. We propose a lightweight synthetic data augmentation pipeline that fine-tunes a LoRA adapter on as few as 20-50 real images of a rare class and uses a pretrained diffusion model to generate synthetic samples for training. We systematically vary the synthetic-to-real ratio and evaluate the approach across two structurally different domains: chest X-ray pathology classification (NIH ChestX-ray14) and industrial surface crack detection (Magnetic Tile Defect dataset). All evaluations are performed on held-out sets of real images only. Across both domains, synthetic augmentation consistently improves rare-class recall and F1 compared to training with real data alone. Performance improves with moderate synthetic augmentation and shows diminishing returns as the synthetic ratio increases. These results suggest that LoRA-adapted diffusion models provide a simple and scalable mechanism for augmenting rare classes, enabling effective learning in data-scarce scenarios across heterogeneous visual domains.

Related papers

• Epidemiology of Model Collapse: Modeling Synthetic Data Contamination via Bilayer SIR Dynamics [1.7832408661523071]
  Training on synthetic data causes model collapse, but existing analyses treat this as single-chain degradation.<n>We propose a bilayer coupled SIR/SIRS framework treating data corpora and AI models as two interacting populations.
  arXiv  Detail & Related papers  (2026-04-14T06:49:30Z)
• Towards Syn-to-Real IQA: A Novel Perspective on Reshaping Synthetic Data Distributions [74.00222571094437]
  Blind Image Quality Assessment (BIQA) has advanced significantly through deep learning, but the scarcity of large-scale labeled datasets remains a challenge.<n>We make a key observation that representations learned from synthetic datasets often exhibit a discrete and clustered pattern that hinders regression performance.<n>We introduce a novel framework SynDR-IQA, which reshapes synthetic data distribution to enhance BIQA generalization.
  arXiv  Detail & Related papers  (2026-01-01T06:11:16Z)
• Towards Application Aligned Synthetic Surgical Image Synthesis [3.1373284090264857]
  We introduce emphSurgical Application-Aligned Diffusion (SAADi), a new framework that aligns diffusion models with samples preferred by downstream models.<n>Our method constructs pairs of emphpreferred and emphnon-preferred synthetic images and employs lightweight fine-tuning of diffusion models to align the image generation process with downstream objectives explicitly.
  arXiv  Detail & Related papers  (2025-09-23T08:40:40Z)
• Improving Heart Rejection Detection in XPCI Images Using Synthetic Data Augmentation [0.0]
  StyleGAN was trained on available 3R biopsy patches and subsequently used to generate 10,000 realistic synthetic images.<n>These were combined with real 0R samples, that is samples without rejection, in various configurations to train ResNet-18 classifiers for binary rejection classification.<n>Results demonstrate that synthetic data improves classification performance, particularly when used in combination with real samples.
  arXiv  Detail & Related papers  (2025-05-26T09:26:36Z)
• SAU: A Dual-Branch Network to Enhance Long-Tailed Recognition via Generative Models [9.340077455871736]
  Long-tailed distributions in image recognition pose a considerable challenge due to the severe imbalance between a few dominant classes. Recently, the use of large generative models to create synthetic data for image classification has been realized. We propose the use of synthetic data as a complement to long-tailed datasets to eliminate the impact of data imbalance.
  arXiv  Detail & Related papers  (2024-08-29T05:33:59Z)
• Training Class-Imbalanced Diffusion Model Via Overlap Optimization [55.96820607533968]
  Diffusion models trained on real-world datasets often yield inferior fidelity for tail classes. Deep generative models, including diffusion models, are biased towards classes with abundant training images. We propose a method based on contrastive learning to minimize the overlap between distributions of synthetic images for different classes.
  arXiv  Detail & Related papers  (2024-02-16T16:47:21Z)
• Derm-T2IM: Harnessing Synthetic Skin Lesion Data via Stable Diffusion Models for Enhanced Skin Disease Classification using ViT and CNN [1.0499611180329804]
  We aim to incorporate enhanced data transformation techniques by extending the recent success of few-shot learning. We investigate the impact of incorporating newly generated synthetic data into the training pipeline of state-of-art machine learning models.
  arXiv  Detail & Related papers  (2024-01-10T13:46:03Z)
• UAV-Sim: NeRF-based Synthetic Data Generation for UAV-based Perception [62.71374902455154]
  We leverage recent advancements in neural rendering to improve static and dynamic novelview UAV-based image rendering. We demonstrate a considerable performance boost when a state-of-the-art detection model is optimized primarily on hybrid sets of real and synthetic data.
  arXiv  Detail & Related papers  (2023-10-25T00:20:37Z)
• Contrastive Model Inversion for Data-Free Knowledge Distillation [60.08025054715192]
  We propose Contrastive Model Inversion, where the data diversity is explicitly modeled as an optimizable objective. Our main observation is that, under the constraint of the same amount of data, higher data diversity usually indicates stronger instance discrimination. Experiments on CIFAR-10, CIFAR-100, and Tiny-ImageNet demonstrate that CMI achieves significantly superior performance when the generated data are used for knowledge distillation.
  arXiv  Detail & Related papers  (2021-05-18T15:13:00Z)
• Adversarial Feature Hallucination Networks for Few-Shot Learning [84.31660118264514]
  Adversarial Feature Hallucination Networks (AFHN) is based on conditional Wasserstein Generative Adversarial networks (cWGAN) Two novel regularizers are incorporated into AFHN to encourage discriminability and diversity of the synthesized features.
  arXiv  Detail & Related papers  (2020-03-30T02:43:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.