可微拉普拉斯矩阵引导的超像素分割
Differentiable Laplacian Matrix Guided Superpixel Segmentation
提供一种模型无关、完全可微的图拉普拉斯损失，可直接嵌入现有超像素网络训练中替代不可微的连通性后处理，做密集预测或区域划分的研究者可复用该正则项与新增的杂散像素/多余连通分量评测指标。
原文依据 We propose a simple, fully differentiable graph-Laplacian loss that encourages spatial regularity and connectivity during training.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论超像素将图像划分为感知上连贯的区域，从而降低下游视觉任务的开销。现代深度学习方法在超像素生成方面表现出色，但往往产生不规则的边界和孤立像素，因此需要借助不可微的后处理来强制满足连通性，这削弱了端到端学习的能力。我们提出一种简单、完全可微的图拉普拉斯损失，在训练过程中促使空间规则性与连通性得到满足。该损失与模型无关，可以无缝集成到现有架构的训练中，以提升超像素的质量。此外，我们引入两个新指标——平均杂散像素数和多余连通分量数——来衡量超像素的质量。我们在强制连通性与不强制连通性两种设定下，均展示了相对于最优方法在定性和定量上的改进。我们的方法朝着消除不可微后处理迈出了重要一步。项目代码：https://github.com/jeremyJJB/Differentiable-Laplacian-Matrix-Guided-Superpixel-Segmentation。
英文原文摘要 1112 字符
Superpixels partition an image into perceptually coherent regions, reducing the cost of downstream vision tasks. Modern deep learning methods excel at superpixel generation but often yield irregular boundaries and isolated pixels, necessitating non-differentiable post-processing to enforce connectivity. This undermines the end-to-end learning capabilities. We propose a simple, fully differentiable graph-Laplacian loss that encourages spatial regularity and connectivity during training. The loss is model-agnostic and can be seamlessly integrated into the training of existing architectures to improve the quality of superpixels. In addition, we introduce two novel metrics, the average stray pixel count and excess component count, to measure the quality of superpixels. We demonstrate both qualitative and quantitative improvements over state-of-the-art methods with and without enforced connectivity. Our approach represents a significant step toward eliminating non-differentiable post-processing. Project code: https://github.com/jeremyJJB/Differentiable-Laplacian-Matrix-Guided-Superpixel-Segmentation.