中文标题#
壓力下的像素：探索高分辨率醫學影像中基礎模型的微調範式
英文标题#
Pixels Under Pressure: Exploring Fine-Tuning Paradigms for Foundation Models in High-Resolution Medical Imaging
中文摘要#
基於擴散的基礎模型的進展提高了文本到圖像的生成，但大多數努力僅限於低分辨率設置。 隨著高分辨率圖像合成在各種應用中變得越來越重要，特別是在醫學成像領域，微調成為適應這些強大的預訓練模型以滿足特定任務需求和數據分佈的關鍵機制。 在本工作中，我們進行了一項系統研究，考察在擴展到高分辨率 512x512 像素時，各種微調技術對圖像生成質量的影響。 我們基準測試了一組多樣化的微調方法，包括完整的微調策略和參數高效的微調（PEFT）。 我們分析了不同的微調方法如何影響關鍵質量指標，包括 Fréchet Inception Distance（FID）、Vendi 分數和提示圖像對齊。 我們還評估了在數據稀缺條件下生成圖像在下游分類任務中的實用性，結果表明，當使用合成圖像進行分類器訓練和在真實圖像上進行評估時，特定的微調策略可以提高生成保真度和下游性能。 我們的代碼可通過項目網站獲取 - https://tehraninasab.github.io/PixelUPressure/.
英文摘要#
Advancements in diffusion-based foundation models have improved text-to-image generation, yet most efforts have been limited to low-resolution settings. As high-resolution image synthesis becomes increasingly essential for various applications, particularly in medical imaging domains, fine-tuning emerges as a crucial mechanism for adapting these powerful pre-trained models to task-specific requirements and data distributions. In this work, we present a systematic study, examining the impact of various fine-tuning techniques on image generation quality when scaling to high resolution 512x512 pixels. We benchmark a diverse set of fine-tuning methods, including full fine-tuning strategies and parameter-efficient fine-tuning (PEFT). We dissect how different fine-tuning methods influence key quality metrics, including Fr'echet Inception Distance (FID), Vendi score, and prompt-image alignment. We also evaluate the utility of generated images in a downstream classification task under data-scarce conditions, demonstrating that specific fine-tuning strategies improve both generation fidelity and downstream performance when synthetic images are used for classifier training and evaluation on real images. Our code is accessible through the project website - https://tehraninasab.github.io/PixelUPressure/.
文章页面#
PDF 获取#
抖音掃碼查看更多精彩內容