提出向量对比学习框架，解决医学图像像素级预训练的特征分散问题。
Vector Contrastive Learning For Pixel-Wise Pretraining In Medical Vision
- 将对比学习重构为向量回归问题，量化像素特征距离。
- 在8项任务中显著提升像素级自监督预训练效果。
- 适合需要保留像素相关性的医学视觉模型研究者使用。
对比学习（CL）已成为基础模型自监督预训练（SSP）的核心方法，但将其扩展到对医学视觉至关重要的像素级表征仍是一大挑战。标准的对比学习将预训练建模为二分类优化问题（二元对比学习），过度追求特征分散导致过分散问题，破坏了像素间的特征相关性，从而扰乱类内分布。本文提出的向量对比学习（Vector CL）将对比学习重构为向量回归问题，通过建模回归位移向量中的特征距离，实现像素级预训练中的分散度量化。为此，我们提出COntrast in VEctor Regression（COVER）框架：建立可扩展的基于向量的自学习机制，确保从向量回归到距离建模的一致优化流程，并采用向量金字塔结构实现粒度自适应，从而在自监督预训练中保持像素级特征的相关性。在涵盖2个维度和4种模态的8项任务上进行的大量实验表明，COVER显著提升了像素级自监督预训练性能，推动了通用医学视觉基础模型的发展。
原文摘要 · Abstract (English)
Contrastive learning (CL) has become a cornerstone of self-supervised pretraining (SSP) in foundation models, however, extending CL to pixel-wise representation, crucial for medical vision, remains an open problem. Standard CL formulates SSP as a binary optimization problem (binary CL) where the excessive pursuit of feature dispersion leads to an over-dispersion problem, breaking pixel-wise feature correlation thus disrupting the intra-class distribution. Our vector CL reformulates CL as a vector regression problem, enabling dispersion quantification in pixel-wise pretraining via modeling feature distances in regressing displacement vectors. To implement this novel paradigm, we propose the COntrast in VEctor Regression (COVER) framework. COVER establishes an extendable vector-based self-learning, enforces a consistent optimization flow from vector regression to distance modeling, and leverages a vector pyramid architecture for granularity adaptation, thus preserving pixel-wise feature correlations in SSP. Extensive experiments across 8 tasks, spanning 2 dimensions and 4 modalities, show that COVER significantly improves pixel-wise SSP, advancing generalizable medical visual foundation models.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。