This paper proposes a pose-graph attentional graph neural network, called P-GAT, which compares (key)nodes between sequential and non-sequential sub-graphs for place recognition tasks as opposed to a common frame-to-frame retrieval problem formulation currently implemented in SOTA place recognition methods. P-GAT uses the maximum spatial and temporal information between neighbour cloud descriptors -- generated by an existing encoder -- utilising the concept of pose-graph SLAM. Leveraging intra- and inter-attention and graph neural network, P-GAT relates point clouds captured in nearby locations in Euclidean space and their embeddings in feature space. Experimental results on the large-scale publically available datasets demonstrate the effectiveness of our approach in scenes lacking distinct features and when training and testing environments have different distributions (domain adaptation). Further, an exhaustive comparison with the state-of-the-art shows improvements in performance gains. Code is available at https://github.com/csiro-robotics/P-GAT.

翻译：本文提出一种名为P-GAT的位姿图注意力图神经网络，该方法通过比较序列子图与非序列子图中的（关键）节点来完成地点识别任务，这与当前最先进地点识别方法中普遍采用的帧对帧检索问题求解范式有所不同。P-GAT利用现有编码器生成的相邻点云描述符之间的最大时空信息，并借鉴位姿图SLAM概念。通过结合内部与交叉注意力机制及图神经网络，P-GAT建立了欧氏空间中邻近位置点云及其特征空间嵌入之间的关联。在公开的大规模数据集上的实验结果表明，该方法在缺乏显著特征场景以及训练与测试环境存在分布差异（域自适应）时均具有有效性。此外，与最先进方法的全面对比显示其在性能提升方面表现优异。代码开源地址：https://github.com/csiro-robotics/P-GAT。