混合标注提升结肠镜图像分割，降低标注成本。
MixPolyp: Integrating Mask, Box and Scribble Supervision for Enhanced Polyp Segmentation
- 融合掩码、边界框和草图三种标注方式，统一训练模型。
- 在五个数据集上均优于单标注方法，显著提升分割精度。
- 适合医疗图像标注资源有限的场景，尤其结肠镜分析。
受限于昂贵的标注成本，结肠息肉分割模型常面临数据不足问题。为此，我们提出混合监督分割范式 MixPolyp。不同于依赖单一标注类型的传统模型，MixPolyp 在同一模型中整合掩码、边界框和草图三类标注，扩大可用数据范围并降低标注成本。为此，提出三种新损失函数：子空间投影损失（L_SP）消除边界框与预测间的形状不一致；二值最小熵损失（L_BME）通过最小熵约束为未标注像素提供监督，缓解标注稀疏性；线性正则化损失（L_LR）强制预测一致性，减少解的非唯一性。这些损失独立于模型结构，仅用于训练，推理无额外开销。在五个数据集上的大量实验表明，MixPolyp 效果显著。
原文摘要 · Abstract (English)
Limited by the expensive labeling, polyp segmentation models are plagued by data shortages. To tackle this, we propose the mixed supervised polyp segmentation paradigm (MixPolyp). Unlike traditional models relying on a single type of annotation, MixPolyp combines diverse annotation types (mask, box, and scribble) within a single model, thereby expanding the range of available data and reducing labeling costs. To achieve this, MixPolyp introduces three novel supervision losses to handle various annotations: Subspace Projection loss (L_SP), Binary Minimum Entropy loss (L_BME), and Linear Regularization loss (L_LR). For box annotations, L_SP eliminates shape inconsistencies between the prediction and the supervision. For scribble annotations, L_BME provides supervision for unlabeled pixels through minimum entropy constraint, thereby alleviating supervision sparsity. Furthermore, L_LR provides dense supervision by enforcing consistency among the predictions, thus reducing the non-uniqueness. These losses are independent of the model structure, making them generally applicable. They are used only during training, adding no computational cost during inference. Extensive experiments on five datasets demonstrate MixPolyp's effectiveness.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。