A Dense Mapping Algorithm Based on Spatiotemporal Consistency

Dense mapping is an important part of mobile robot navigation and environmental understanding. Aiming to address the problem that Dense Surfel Mapping relies on the input of a common-view relationship, we propose a local map extraction strategy based on spatiotemporal consistency. The local map is extracted through the inter-frame pose observability and temporal continuity. To reduce the blurring of map fusion caused by the different viewing angles, a normal constraint is added to the map fusion and weight initialization. To achieve continuous and stable time efficiency, we dynamically adjust the parameters of superpixel extraction. The experimental results on the ICL-NUIM and KITTI datasets show that the partial reconstruction accuracy is improved by approximately 27-43%. In addition, the system achieves a greater than 15 Hz real-time performance using only CPU computation, which is improved by approximately 13%.

Automated summary

In this paper, a local map extraction strategy based on spatiotemporal consistency is proposed to address the problem that Dense Surfel Mapping relies on the input of a common-view relationship. The local map is extracted through the inter-frame pose observability and temporal continuity. A normal constraint is added to the map fusion and weight initialization to reduce the blurring caused by different viewing angles. The parameters of superpixel extraction are dynamically adjusted to achieve continuous and stable time efficiency. Experimental results on the ICL-NUIM and KITTI datasets show an improvement of approximately 27-43% in partial reconstruction accuracy. Furthermore, the system achieves a real-time performance of greater than 15 Hz using only CPU computation, with an improvement of approximately 13%.