PromptMoE：利用混合专家改进提示的分割细化框架
PromptMoE: A Segmentation Refinement Framework Leveraging Mixture of Experts for Improved Prompting
把 MoE 与动态专家选择引入提示词生成环节，用图像+粗掩码预测专家得分图并自适应放置点提示词，可作为即插即用的分割细化模块，替换 CascadePSP/SegRefiner 类固定提示规则，对医学、质检等边界敏感场景的掩码后处理有直接复用价值。
原文依据 We propose PromptMoE, a model-agnostic MoE-driven prompting refiner effective in segmentation refinement across tasks and domains.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论在边界精度至关重要的视觉任务（如医学诊断、质量控制等）中，高质量分割十分关键。近期，可提示视觉模型已成为分割细化框架的有效骨干。然而，其性能不仅取决于提示词质量，还必须克服可提示模型输出的噪声输入掩码与语义歧义结果。现有基于提示的细化器依赖固定的提示规则，因而对变化的失效模式以及新任务或新领域十分脆弱。我们提出 PromptMoE，一个模型无关、由混合专家（MoE）驱动的提示细化器，可跨任务与跨领域有效地完成分割细化。PromptMoE 包含三个协同模块来细化初始掩码：基于 MoE 的图像信息提示框架（IIP）接收图像与粗掩码，生成一组专家得分图以指导提示词生成；动态专家选择器（DES）仅激活最相关的专家并融合其得分图，以避免稠密评估与信号稀释；提示放置探索器（PPE）在融合后的引导图上进行探索，以放置高置信度、空间多样的点提示词。在五个基准数据集（BIG、VOC、DAVIS585、ECSSD、MSRA-B）上，PromptMoE 在语义、实例与显著性任务上相较 SOTA 方法 CascadePSP、SegRefiner 与 SAMRefiner 取得统计显著提升，平均提升 +6.24 IoU / +8.99 BIoU。
英文原文摘要 1449 字符
High-quality segmentations are critical in vision tasks where boundary accuracy is important (e.g., medical diagnostics, quality control, etc.). Recently, promptable vision models have emerged as effective backbones for segmentation refinement frameworks. However, their performance not only hinges on prompt quality, they also must overcome noisy input masks and semantically ambiguous outputs from promptable models. Existing prompt-based refiners rely on fixed prompt rules, making them brittle to changing failure modes and new tasks or domains. We propose PromptMoE, a model-agnostic MoE-driven prompting refiner effective in segmentation refinement across tasks and domains. PromptMoE features three collaborative modules to refine an initial mask: our MoE-based Image-Informed Prompting framework (IIP) takes an image and coarse mask and produces a set of expert score maps to guide prompt generation, the Dynamic Expert Selector (DES) activates only the most relevant experts and fuses their maps to avoid dense evaluation and signal dilution, and the Prompt-Placement Explorer (PPE) explores the fused guidance map to place high-confidence spatially diverse point prompts. Across five benchmark datasets (BIG, VOC, DAVIS585, ECSSD, MSRA-B), PromptMoE achieves statistically significant gains over SOTA methods CascadePSP, SegRefiner, and SAMRefiner on semantic, instance, and salient tasks, with mean improvements of +6.24 IoU / +8.99 BIoU.