arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.06589cs.CV
收藏

Seg-HGNN:基于双曲图神经网络的无监督轻量级图像分割

Seg-HGNN: Unsupervised and Light-Weight Image Segmentation with Hyperbolic Graph Neural Networks

Samsung R&D Institute India(三星印度研发中心)

Debjyoti Mondal, Rahul Mishra, Chandan Pandey

AI总结:

本文提出Seg-HGNN,一种基于双曲图神经网络的无监督轻量级图像分割方法,以极小的嵌入尺寸和少于7.5k参数在多个基准上超越现有最佳无监督方法,并实现快速推理。

摘要:

通过线性超空间在欧氏空间中进行图像分析已被广泛研究。然而,为了寻求更有效的图像表示,我们转向双曲流形。它们提供了一种引人注目的替代方案,能够以极小的维度捕捉图像中复杂的层次关系。为了证明双曲嵌入的能力,我们引入了一种用于图像分割的轻量级双曲图神经网络,以非常小的嵌入尺寸包含块级特征。我们的解决方案Seg-HGNN在VOC-07、VOC-12上的定位任务中分别比当前最佳无监督方法高出2.5%和4%,在CUB-200、ECSSD上的分割任务中分别高出0.8%和1.3%。Seg-HGNN的可训练参数少于7.5k,在GTX1650等非常标准的GPU上提供有效且快速(约2张图像/秒)的结果。该实证评估为双曲表示在视觉任务中的有效性和潜力提供了令人信服的证据。

英文摘要:

Image analysis in the euclidean space through linear hyperspaces is well studied. However, in the quest for more effective image representations, we turn to hyperbolic manifolds. They provide a compelling alternative to capture complex hierarchical relationships in images with remarkably small dimensionality. To demonstrate hyperbolic embeddings' competence, we introduce a light-weight hyperbolic graph neural network for image segmentation, encompassing patch-level features in a very small embedding size. Our solution, Seg-HGNN, surpasses the current best unsupervised method by 2.5\%, 4\% on VOC-07, VOC-12 for localization, and by 0.8\%, 1.3\% on CUB-200, ECSSD for segmentation, respectively. With less than 7.5k trainable parameters, Seg-HGNN delivers effective and fast ($\approx 2$ images/second) results on very standard GPUs like the GTX1650. This empirical evaluation presents compelling evidence of the efficacy and potential of hyperbolic representations for vision tasks.

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。

↑