发表机构
University of Waterloo（滑铁卢大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
SPARC提出一种基于超像素的区域级对比学习框架，通过区域对比分支与全局目标联合优化，在语义分割和检测上显著超越现有方法，提升密集预测性能。
AI 中文摘要
自监督学习（SSL）已成为无需人工标注即可学习视觉表示的有效方法。在自监督学习方法中，对比学习被广泛用于视觉表示学习。然而，现有的对比自监督方法主要关注图像级或像素级的表示学习，而区域级的表示学习仍较少被探索。我们提出了SPARC，一种区域级对比学习框架，利用超像素在增强图像视图之间建立显式对应关系。SPARC引入了一个区域对比分支，该分支执行基于超像素的特征池化，并将区域级对比目标与全局图像级目标联合优化。在相同设置下，SPARC持续优于先前的方法，如MoCo-v2和DenseCL，在语义分割上实现了高达+9.79 mIoU的提升，在目标检测上实现了+4.88 AP的提升。消融研究进一步表明，区域级目标产生了最强的性能。因此，区域级对比学习是改进自监督视觉预训练以用于密集预测任务的有效方法。代码仓库可通过此https URL访问。
英文摘要
Self-supervised learning (SSL) has become an effective approach for learning visual representations without manual annotations. Among SSL approaches, contrastive learning has been widely used for visual representation learning. However, existing contrastive SSL methods have focused primarily on image-level or pixel-level representation learning, while region-level representation learning remains less explored. We propose SPARC, a region-level contrastive learning framework that leverages superpixels to establish explicit correspondence between augmented image views. SPARC introduces a region contrastive branch that performs superpixel-based feature pooling and optimizes a region-level contrastive objective jointly with a global image-level objective. Under identical settings, SPARC consistently outperforms previous methods such as MoCo-v2 and DenseCL, achieving improvements of up to +9.79 mIoU for semantic segmentation and +4.88 AP for object detection. Ablation studies further demonstrate that region-level objectives produce the strongest performance. Thus, region-level contrastive learning is an effective approach for improving self-supervised visual pretraining for dense prediction tasks. Code repository can be accessed at https://github.com/xRIPEIx/SPARC.
Comments5 pages, 2 figures. Submitted to the IEEE ICASSP 2027 Conference
