Vision-and-language navigation (VLN) enables the agent to navigate to a remote location following the natural language instruction in 3D environments. To represent the previously visited environment, most approaches for VLN implement memory using recurrent states, topological maps, or top-down semantic maps. In contrast to these approaches, we build the top-down egocentric and dynamically growing Grid Memory Map (i.e., GridMM) to structure the visited environment. From a global perspective, historical observations are projected into a unified grid map in a top-down view, which can better represent the spatial relations of the environment. From a local perspective, we further propose an instruction relevance aggregation method to capture fine-grained visual clues in each grid region. Extensive experiments are conducted on both the REVERIE, R2R, SOON datasets in the discrete environments, and the R2R-CE dataset in the continuous environments, showing the superiority of our proposed method.
@article{arxiv.2307.12907,
title = {GridMM: Grid Memory Map for Vision-and-Language Navigation},
author = {Zihan Wang and Xiangyang Li and Jiahao Yang and Yeqi Liu and Shuqiang Jiang},
journal= {arXiv preprint arXiv:2307.12907},
year = {2023}
}
Comments
Accepted by ICCV 2023. The code is available at https://github.com/MrZihan/GridMM