Many challenges from natural world can be formulated as a graph matching problem. Previous deep learning-based methods mainly consider a full two-graph matching setting. In this work, we study the more general partial matching problem with multi-graph cycle consistency guarantees. Building on a recent progress in deep learning on graphs, we propose a novel data-driven method (URL) for partial multi-graph matching, which uses an object-to-universe formulation and learns latent representations of abstract universe points. The proposed approach advances the state of the art in semantic keypoint matching problem, evaluated on Pascal VOC, CUB, and Willow datasets. Moreover, the set of controlled experiments on a synthetic graph matching dataset demonstrates the scalability of our method to graphs with large number of nodes and its robustness to high partiality.
translated by 谷歌翻译
我们解决了最小化一类能量功能的问题,该功能由数据和平滑度术语组成,这些术语通常发生在机器学习,计算机视觉和模式识别中。尽管离散优化方法能够提供理论最优保证,但它们只能处理有限数量的标签,因此会遭受标签离散偏置的困扰。现有的连续优化方法可以找到Sublabel精确的解决方案,但对于大型标签空间而言,它们并不有效。在这项工作中,我们提出了一种有效的Sublabel精确方法,该方法利用了连续模型和离散模型的最佳属性。我们将问题分为两个顺序的步骤:(i)选择标签范围的全局离散优化,以及(ii)在所选范围内的能量函数凸的有效连续连续的sublabel-carcurate局部改进。这样做可以使我们能够提高时间和记忆效率,同时实际上将准确性保持在与连续凸放放松方法相同的水平上,此外,在离散方法级别上提供了理论最佳保证。最后,我们显示了提出的对一般成对平滑度项的拟议方法的灵活性,因此它适用于广泛的正则化。图像授予问题的说明示例的实验证明了该方法的特性。代码复制实验可在\ url {https://github.com/nurlanov-zh/sublabel-accurate-alpha-expansion}获得。
translated by 谷歌翻译