CCTNet: A Circular Convolutional Transformer Network for LiDAR-based Place Recognition Handling Movable Objects Occlusion

Place recognition is a fundamental task in robotics, enabling loop closure detection in simultaneous localization and mapping (SLAM), and re-localization on prior maps. Current range image-based networks use single-column convolution to maintain feature invariance to shifts in image columns caused by light detection and ranging (LiDAR) viewpoint change. However, this raises the issues such as “restricted receptive fields” and “excessive focus on local regions”, degrading the performance of networks, especially in scenarios with movable objects. In this paper, a lightweight circular convolutional Transformer network named CCTNet is proposed, which aims to boost performance by capturing structural information in point clouds and facilitating cross-dimensional interaction of spatial and channel information. Initially, a Circular Convolution Module (CCM) is introduced, expanding the network’s perceptual field while maintaining feature consistency across varying LiDAR perspectives. Then, a Range Transformer Module (RTM) is proposed, which enhances place recognition accuracy in scenarios with movable objects by employing a combination of channel and spatial attention mechanisms. Furthermore, we propose an overlap-based loss function, transforming the place recognition task from a binary loop closure classification into a regression problem linked to the overlap between LiDAR frames. Through extensive experiments on the Karlsruhe Institute of Technology and Toyota Technological Institute at Chicago (KITTI) and Ford Campus datasets, CCTNet surpasses comparable methods, achieving Recall@1 of 0.924 and 0.965, and Recall@1% of 0.990 and 0.993 on the test set, showcasing superior performance. Results on the self-collected dataset further demonstrate the proposed method’s potential for practical implementation in complex scenarios to handle movable objects, showing improved generalization across various datasets.