Scene graph generation (SGG) aims to detect objects and predict the relationships between each pair of objects. Existing SGG methods usually suffer from several issues, including 1) ambiguous object representations, as graph neural network-based message passing (GMP) modules are typically sensitive to spurious inter-node correlations, and 2) low diversity in relationship predictions due to severe class imbalance and a large number of missing annotations. To address both problems, in this paper, we propose a regularized unrolling network (RU-Net). We first study the relation between GMP and graph Laplacian denoising (GLD) from the perspective of the unrolling technique, determining that GMP can be formulated as a solver for GLD. Based on this observation, we propose an unrolled message passing module and introduce an $\ell_p$-based graph regularization to suppress spurious connections between nodes. Second, we propose a group diversity enhancement module that promotes the prediction diversity of relationships via rank maximization. Systematic experiments demonstrate that RU-Net is effective under a variety of settings and metrics. Furthermore, RU-Net achieves new state-of-the-arts on three popular databases: VG, VRD, and OI. Code is available at https://github.com/siml3/RU-Net.

翻译：光谱图生成(SGG)旨在探测对象并预测每对对象之间的关系。现有的SGG方法通常存在若干问题,包括:(1) 模糊的物体表示,因为基于图形神经网络的电文传递模块通常对虚假的跨节点相互关系十分敏感;(2) 由于严重的阶级不平衡和大量缺失的注释,关系预测的多样性较低。为了解决这两个问题,我们在本文件中建议建立一个正规化的无滚动网络(RU-Net)。我们首先从无滚动技术的角度研究GMP与Laplacian图的脱色(GLD)之间的关系,确定GMP可被设计成GLD的解答器。基于这一观察,我们提出了一个无滚动的信息传递模块,并引入一个$@ell_p$基础的图形规范,以压制节点之间的虚假联系。第二,我们提议了一个通过定级最大化促进预测关系多样性的群集模块。系统实验表明,RU-Net在各种设置和计量仪下是有效的,确定GPPPP可以被设计成为G的。此外,RU-Net在新的州数据库中,O-RB/GRP/Comms。