发表机构
Télécom Paris, Institut Polytechnique de Paris; Hôpital Universitaire Lariboisière (AP-HP); Université Paris Cité; Inserm（巴黎综合理工学院电信学院; 拉里布瓦西耶尔大学医院（巴黎公立医院集团）; 巴黎西岱大学; 法国国家健康与医学研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
Pix2Rep-v2是一种密集自监督学习框架，通过像素级冗余减少与等变原理提升数据效率，在少样本医学影像任务中优于全监督基线，可替代线性探测或全微调方法。
AI 中文摘要
密集自监督学习（SSL）是一种无需标注即可学习解决密集医学影像任务所需局部描述子的强大范式。我们提出Pix2Rep-v2，一种适用于少样本下游应用的像素级和体素级表示的SSL框架。Pix2Rep-v2通过利用像素级的冗余减少目标，结合密集表示的等变原理，解决了密集SSL的主要挑战，可高效扩展至3D或宽视场应用。我们在四个数据集、多个任务、多种模态、多种解剖结构上，使用2D和3D的多种骨干网络，并在不同数据 regime 下评估了我们的方法。作为下游任务线性探测或全微调的替代方案，我们还提出了一种基于密集原型方法的上下文内变体，无需下游训练。Pix2Rep-v2在少样本场景下表现出比全监督基线显著更高的数据效率，且与当前最优方法具有竞争力，例如在M&Ms-2数据集的一次分割任务中Dice分数提升9.3个百分点。我们的代码和预训练模型可在该https URL公开获取。
英文摘要
Dense self-supervised learning (SSL) is a powerful paradigm for learning without annotations the local descriptors required to solve dense medical imaging tasks. We present Pix2Rep-v2, a framework for SSL of pixel- and voxel-level representations suitable for few-shot downstream applications. Pix2Rep-v2 addresses the main challenges of dense SSL by leveraging a redundancy reduction objective at the pixel-level with a principle of equivariance of dense representations, that scales efficiently to 3D or wide field-of-view applications. We evaluate our method on four datasets, across multiple tasks, multiple modalities and anatomical structures using multiple backbones in 2D and 3D, and under various data regimes. As an alternative to linear probing or full fine-tuning on the downstream task, we also propose an in-context variant, without downstream training, based on a dense prototype approach. Pix2Rep-v2 shows substantially higher data-efficiency in few-shot scenarios compared to fully supervised baselines, and is competitive with the state-of-the-art e.g., +9.3 Dice points in one-shot segmentation on the M&Ms-2 dataset. Our code and pre-trained models are publicly available at https://github.com/BioMedTP/pix2rep-v2.
CommentsAccepted at MICCAI 2026
