Improving Visual Place Recognition with Sequence-Matching Receptiveness Prediction
This work addresses the challenge of unreliable sequence matching in VPR for robotics and autonomous systems, offering an incremental improvement by making existing methods more robust.
The paper tackles the problem of unpredictable performance in visual place recognition (VPR) when using sequence-based matching by introducing a supervised learning approach that predicts per-frame sequence matching receptiveness, enabling selective trust in sequence matching outputs and significantly improving VPR performance across multiple state-of-the-art techniques and three benchmark datasets.
In visual place recognition (VPR), filtering and sequence-based matching approaches can improve performance by integrating temporal information across image sequences, especially in challenging conditions. While these methods are commonly applied, their effects on system behavior can be unpredictable and can actually make performance worse in certain situations. In this work, we present a new supervised learning approach that learns to predict the per-frame sequence matching receptiveness (SMR) of VPR techniques, enabling the system to selectively decide when to trust the output of a sequence matching system. Our approach is agnostic to the underlying VPR technique and effectively predicts SMR, and hence significantly improves VPR performance across a large range of state-of-the-art and classical VPR techniques (namely CosPlace, MixVPR, EigenPlaces, SALAD, AP-GeM, NetVLAD and SAD), and across three benchmark VPR datasets (Nordland, Oxford RobotCar, and SFU-Mountain). We also provide insights into a complementary approach that uses the predictor to replace discarded matches, and present ablation studies including an analysis of the interactions between our SMR predictor and the selected sequence length.