TOSA

Abstract

Despite advancements in visual Simultaneous Localization and Mapping (SLAM), prevailing visual Loop Closure Detection (LCD) methods primarily rely on computationally intensive image similarity comparisons, neglecting temporal-spatial context during long-term exploration. To address this issue, we propose TOSA, a novel vLCD algorithm harnessing TempOral and SpAtial context for efficient LCD. Specifically, as the agent explores through time, our approach recurrently updates a latent feature incorporating historical information via a Long Short-Term Memory (LSTM) module. Upon receiving a query frame, TOSA seamlessly fuses the latent feature with the query feature to predict the candidates’ distribution, thus averting intensive similarity computation. Additionally, TOSA integrates a temporal-spatial convolution for candidate refinement by thoroughly exploiting the temporal consistency and spatial correlation to enhance selected candidates, further boosting the performance. Extensive experiments across four standard datasets showcase the superiority of our method over existing state-of-the-art techniques, demonstrating the effectiveness of utilizing rich temporal-spatial contexts.

Pipeline

(a) We select N loop candidates for the query frame by leveraging temporal context. (b) We refine the loop candidates using a novel temporal-spatial convolution operation.

Experiments

TOSA performs best in terms of efficacy and can achieve real-time efficiency.

Citation


        @inproceedings{li2024visual,
        title={Visual Loop Closure Detection with Thorough Temporal and Spatial Context Exploitation},
        author={Li, Jiaxin and Wang, Zan and Di, Huijun and Li, Jian and Liang, Wei},
        booktitle={International Conference on Intelligent Robots and Systems (IROS)},
        year={2024}
      }