Diffusion Curriculum: Synthetic-to-Real Generative Curriculum Learning via Image-Guided Diffusion
- URL: http://arxiv.org/abs/2410.13674v2
- Date: Fri, 18 Oct 2024 03:28:38 GMT
- Title: Diffusion Curriculum: Synthetic-to-Real Generative Curriculum Learning via Image-Guided Diffusion
- Authors: Yijun Liang, Shweta Bhardwaj, Tianyi Zhou,
- Abstract summary: Low-quality or scarce data has posed significant challenges for training deep neural networks in practice.
Diffusion Curriculum (DisCL) adjusts the image guidance level of image synthesis for each training stage.
DisCL focuses on lower-guidance images of high-quality to learn features as a warm-up of learning higher-guidance images that might be weak on diversity or quality.
- Score: 19.54496184675988
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Low-quality or scarce data has posed significant challenges for training deep neural networks in practice. While classical data augmentation cannot contribute very different new data, diffusion models opens up a new door to build self-evolving AI by generating high-quality and diverse synthetic data through text-guided prompts. However, text-only guidance cannot control synthetic images' proximity to the original images, resulting in out-of-distribution data detrimental to the model performance. To overcome the limitation, we study image guidance to achieve a spectrum of interpolations between synthetic and real images. With stronger image guidance, the generated images are similar to the training data but hard to learn. While with weaker image guidance, the synthetic images will be easier for model but contribute to a larger distribution gap with the original data. The generated full spectrum of data enables us to build a novel "Diffusion Curriculum (DisCL)". DisCL adjusts the image guidance level of image synthesis for each training stage: It identifies and focuses on hard samples for the model and assesses the most effective guidance level of synthetic images to improve hard data learning. We apply DisCL to two challenging tasks: long-tail (LT) classification and learning from low-quality data. It focuses on lower-guidance images of high-quality to learn prototypical features as a warm-up of learning higher-guidance images that might be weak on diversity or quality. Extensive experiments showcase a gain of 2.7% and 2.1% in OOD and ID macro-accuracy when applying DisCL to iWildCam dataset. On ImageNet-LT, DisCL improves the base model's tail-class accuracy from 4.4% to 23.64% and leads to a 4.02% improvement in all-class accuracy.
Related papers
- SkinDualGen: Prompt-Driven Diffusion for Simultaneous Image-Mask Generation in Skin Lesions [0.0]
We propose a novel method that leverages the pretrained Stable Diffusion-2.0 model to generate high-quality synthetic skin lesion images.<n>A hybrid dataset combining real and synthetic data markedly enhances the performance of classification and segmentation models.
arXiv Detail & Related papers (2025-07-26T15:00:37Z) - Stylized Structural Patterns for Improved Neural Network Pre-training [1.8641315013048299]
Deep learning models in computer vision require large datasets of real images, which are difficult to curate and pose privacy and legal concerns.<n>Recent works suggest synthetic data as an alternative, yet models trained with it often underperform.<n>We propose an improved neural fractal formulation through which we introduce a new class of synthetic data.<n>Second, we propose reverse stylization, a technique that transfers visual features from a small, license-free set of real images onto synthetic datasets.
arXiv Detail & Related papers (2025-06-24T09:47:31Z) - LoFT: LoRA-fused Training Dataset Generation with Few-shot Guidance [96.6544564242316]
We introduce a novel dataset generation framework named LoFT, LoRA-Fused Training-data Generation with Few-shot Guidance.<n>Our method fine-tunes LoRA weights on individual real images and fuses them at inference time, producing synthetic images that combine the features of real images for improved diversity and fidelity of generated data.<n>Our experiments show that training on LoFT-generated data consistently outperforms other synthetic dataset methods, significantly increasing accuracy as the dataset size increases.
arXiv Detail & Related papers (2025-05-16T21:17:55Z) - Your Image Generator Is Your New Private Dataset [4.09225917049674]
Generative diffusion models have emerged as powerful tools to synthetically produce training data.
This paper proposes the Text-Conditioned Knowledge Recycling pipeline to tackle these challenges.
The pipeline is rigorously evaluated on ten diverse image classification benchmarks.
arXiv Detail & Related papers (2025-04-06T18:46:08Z) - Gen-SIS: Generative Self-augmentation Improves Self-supervised Learning [52.170253590364545]
Gen-SIS is a diffusion-based augmentation technique trained exclusively on unlabeled image data.
We show that these self-augmentations', i.e. generative augmentations based on the vanilla SSL encoder embeddings, facilitate the training of a stronger SSL encoder.
arXiv Detail & Related papers (2024-12-02T16:20:59Z) - Analysis of Classifier Training on Synthetic Data for Cross-Domain Datasets [4.696575161583618]
This study focuses on camera-based traffic sign recognition applications for advanced driver assistance systems and autonomous driving.
The proposed augmentation pipeline of synthetic datasets includes novel augmentation processes such as structured shadows and gaussian specular highlights.
Experiments showed that a synthetic image-based approach outperforms in most cases real image-based training when applied to cross-domain test datasets.
arXiv Detail & Related papers (2024-10-30T07:11:41Z) - Is Synthetic Image Useful for Transfer Learning? An Investigation into Data Generation, Volume, and Utilization [62.157627519792946]
We introduce a novel framework called bridged transfer, which initially employs synthetic images for fine-tuning a pre-trained model to improve its transferability.
We propose dataset style inversion strategy to improve the stylistic alignment between synthetic and real images.
Our proposed methods are evaluated across 10 different datasets and 5 distinct models, demonstrating consistent improvements.
arXiv Detail & Related papers (2024-03-28T22:25:05Z) - 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) - Learned representation-guided diffusion models for large-image generation [58.192263311786824]
We introduce a novel approach that trains diffusion models conditioned on embeddings from self-supervised learning (SSL)
Our diffusion models successfully project these features back to high-quality histopathology and remote sensing images.
Augmenting real data by generating variations of real images improves downstream accuracy for patch-level and larger, image-scale classification tasks.
arXiv Detail & Related papers (2023-12-12T14:45:45Z) - Scaling Laws of Synthetic Images for Model Training ... for Now [54.43596959598466]
We study the scaling laws of synthetic images generated by state of the art text-to-image models.
We observe that synthetic images demonstrate a scaling trend similar to, but slightly less effective than, real images in CLIP training.
arXiv Detail & Related papers (2023-12-07T18:59:59Z) - Improving the Effectiveness of Deep Generative Data [5.856292656853396]
Training a model on purely synthetic images for downstream image processing tasks results in an undesired performance drop compared to training on real data.
We propose a new taxonomy to describe factors contributing to this commonly observed phenomenon and investigate it on the popular CIFAR-10 dataset.
Our method outperforms baselines on downstream classification tasks both in case of training on synthetic only (Synthetic-to-Real) and training on a mix of real and synthetic data.
arXiv Detail & Related papers (2023-11-07T12:57:58Z) - Image Captions are Natural Prompts for Text-to-Image Models [70.30915140413383]
We analyze the relationship between the training effect of synthetic data and the synthetic data distribution induced by prompts.
We propose a simple yet effective method that prompts text-to-image generative models to synthesize more informative and diverse training data.
Our method significantly improves the performance of models trained on synthetic training data.
arXiv Detail & Related papers (2023-07-17T14:38:11Z) - Performance of GAN-based augmentation for deep learning COVID-19 image
classification [57.1795052451257]
The biggest challenge in the application of deep learning to the medical domain is the availability of training data.
Data augmentation is a typical methodology used in machine learning when confronted with a limited data set.
In this work, a StyleGAN2-ADA model of Generative Adversarial Networks is trained on the limited COVID-19 chest X-ray image set.
arXiv Detail & Related papers (2023-04-18T15:39:58Z) - CIFAKE: Image Classification and Explainable Identification of
AI-Generated Synthetic Images [7.868449549351487]
This article proposes to enhance our ability to recognise AI-generated images through computer vision.
The two sets of data present as a binary classification problem with regard to whether the photograph is real or generated by AI.
This study proposes the use of a Convolutional Neural Network (CNN) to classify the images into two categories; Real or Fake.
arXiv Detail & Related papers (2023-03-24T16:33:06Z)
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.