用于医学视觉像素级预训练的向量对比学习
Vector Contrastive Learning For Pixel-Wise Pretraining In Medical Vision
AI总结：
针对像素级对比学习中的过度离散问题，提出向量回归对比框架COVER，通过建模位移向量量化特征离散，显著提升了医学视觉基础模型的像素级预训练效果。
摘要：
对比学习（CL）已成为基础模型中自监督预训练（SSP）的基石，然而，将CL扩展至对医学视觉至关重要的像素级表示仍是一个悬而未决的问题。标准CL将SSP构建为二元优化问题（二元CL），其中对特征离散的过度追求会导致过度离散问题，破坏像素级特征相关性，从而扰乱类内分布。我们的向量CL将CL重新构建为向量回归问题，通过在回归位移向量中对特征距离进行建模，实现像素级预训练中的离散量化。为实施这一新范式，我们提出了向量回归对比（COVER）框架。COVER建立了一个可扩展的基于向量的自学习机制，强制执行从向量回归到距离建模的一致优化流，并利用向量金字塔架构进行粒度适应，从而在SSP中保留像素级特征相关性。跨越8项任务、涵盖2个维度和4种模态的广泛实验表明，COVER显著改善了像素级SSP，推动了泛化医学视觉基础模型的发展。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
