arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.09893cs.CV
收藏

解决多数据集图像分割中的语义不一致问题

Resolving Inconsistent Semantics in Multi-Dataset Image Segmentation

Rutgers University(罗格斯大学);NEC Laboratories America(美国NEC实验室)

Qilong Zhangli, Di Liu, Abhishek Aich, Dimitris Metaxas, Samuel Schulter

AI总结:

针对多数据集合并训练图像分割模型时的语义不一致问题,提出融合类别语言嵌入与标签空间特定查询嵌入的训练方法,在多类分割任务及新指标上均优于现有方法。

摘要:

利用多个训练数据集来扩展图像分割模型的规模,有助于提升模型的鲁棒性和语义理解能力。单个数据集都有定义明确的真值,其掩码布局互不重叠,语义也相互排斥。然而,将这些数据集合并用于多数据集训练会打破这种协调性,引发语义不一致问题;例如,一个数据集中的“person(人)”类别和另一个数据集中的“face(人脸)”类别,会要求对某些像素进行多标签处理。现有方法在这种设置下表现不佳,尤其是在使用由各训练集混合而成的标签空间进行评估时。为解决这些问题,我们提出了一种简单却有效的多数据集训练方法,该方法整合了类别的基于语言的嵌入和特定于标签空间的查询嵌入。我们的方法无论训练数据集之间存在怎样的底层不一致性,都能保持高性能。值得注意的是,在推理阶段存在标签空间不一致的四个基准数据集上,我们的方法在语义分割任务上的mIoU比现有方法高1.6%,全景分割的PQ高9.1%,实例分割的AP高12.1%,在新提出的PIQ指标上高3.0%。

英文摘要:

Leveraging multiple training datasets to scale up image segmentation models is beneficial for increasing robustness and semantic understanding. Individual datasets have well-defined ground truth with non-overlapping mask layouts and mutually exclusive semantics. However, merging them for multi-dataset training disrupts this harmony and leads to semantic inconsistencies; for example, the class "person" in one dataset and class "face" in another will require multilabel handling for certain pixels. Existing methods struggle with this setting, particularly when evaluated on label spaces mixed from the individual training sets. To overcome these issues, we introduce a simple yet effective multi-dataset training approach by integrating language-based embeddings of class names and label space-specific query embeddings. Our method maintains high performance regardless of the underlying inconsistencies between training datasets. Notably, on four benchmark datasets with label space inconsistencies during inference, we outperform previous methods by 1.6% mIoU for semantic segmentation, 9.1% PQ for panoptic segmentation, 12.1% AP for instance segmentation, and 3.0% in the newly proposed PIQ metric.

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。

↑