Visual Loop Closure Detection with Thorough Temporal and Spatial Context Exploitation

Despite advancements in visual Simultaneous Lo- calization and Mapping (SLAM), prevailing visual Loop Clo- sure Detection (LCD) methods primarily rely on compu- tationally intensive image similarity comparisons, neglecting temporal-spatial context during long-term exploration. To ad- dress this issue, we propose TOSA, a novel visual LCD 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 candi- dates, 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.