无监督通用图像分割
Unsupervised Universal Image Segmentation
AI总结：
提出U2Seg，一种基于自监督聚类生成伪标签并自训练的统一框架，同时处理实例、语义和全景分割，在多个任务上超越现有专门方法，并建立无监督全景分割新基线。
摘要：
已有若干无监督图像分割方法被提出，这些方法消除了对密集人工标注分割掩码的需求；当前模型分别处理语义分割（如STEGO）或类无关实例分割（如CutLER），但不同时处理两者（即全景分割）。我们提出了一种无监督通用分割模型（U2Seg），擅长使用一种新颖的统一框架执行各种图像分割任务——实例、语义和全景分割。U2Seg通过利用自监督模型后进行聚类，为这些分割任务生成伪语义标签；每个聚类代表像素的不同语义和/或实例成员关系。然后我们在这些伪语义标签上对模型进行自训练，相比针对每项任务特化的方法取得了显著的性能提升：在COCO上的无监督实例分割中，相比CutLER，AP$^{\ ext{box}}$提升了+2.6；在COCOStuff上的无监督语义分割中，相比STEGO，PixelAcc提升了+7.0。此外，我们的方法为无监督全景分割设立了新基线，此前该任务尚未被探索。U2Seg也是少样本分割的强预训练模型，在低数据场景（例如仅使用1%的COCO标签）下训练时，超越CutLER达+5.0 AP$^{\ ext{mask}}$。我们希望这种简单而有效的方法能激发更多关于无监督通用图像分割的研究。
英文摘要：
Several unsupervised image segmentation approaches have been proposed which eliminate the need for dense manually-annotated segmentation masks; current models separately handle either semantic segmentation (e.g., STEGO) or class-agnostic instance segmentation (e.g., CutLER), but not both (i.e., panoptic segmentation). We propose an Unsupervised Universal Segmentation model (U2Seg) adept at performing various image segmentation tasks -- instance, semantic and panoptic -- using a novel unified framework. U2Seg generates pseudo semantic labels for these segmentation tasks via leveraging self-supervised models followed by clustering; each cluster represents different semantic and/or instance membership of pixels. We then self-train the model on these pseudo semantic labels, yielding substantial performance gains over specialized methods tailored to each task: a +2.6 AP$^{\text{box}}$ boost vs. CutLER in unsupervised instance segmentation on COCO and a +7.0 PixelAcc increase (vs. STEGO) in unsupervised semantic segmentation on COCOStuff. Moreover, our method sets up a new baseline for unsupervised panoptic segmentation, which has not been previously explored. U2Seg is also a strong pretrained model for few-shot segmentation, surpassing CutLER by +5.0 AP$^{\text{mask}}$ when trained on a low-data regime, e.g., only 1% COCO labels. We hope our simple yet effective method can inspire more research on unsupervised universal image segmentation.
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。
