Search papers, labs, and topics across Lattice.
This paper introduces a novel place-wise dataset pruning framework specifically designed for visual place recognition (VPR), addressing the inefficiencies of traditional sample-wise pruning methods. By employing two new metrics鈥攊ntra-place diversity (IPD) and inter-place similarity (IPS)鈥攖he approach evaluates the training value of entire places rather than individual images, resulting in a more informative and compact dataset. Experimental results show that this method significantly outperforms existing state-of-the-art dataset pruning techniques while drastically reducing both selection and training costs, achieving competitive recognition performance on benchmark datasets.
Pruning entire places instead of individual images can enhance dataset efficiency and model performance in visual place recognition by up to 3.5 times.
Recent visual place recognition (VPR) studies have increasingly relied on large-scale datasets to train more robust and discriminative models. Although this trend significantly improves recognition performance, it also introduces substantial storage and training costs, especially when new architectures or training strategies need to be repeatedly developed and evaluated. Dataset pruning (DP) provides a promising way to improve data efficiency by retaining only informative training data. However, conventional DP methods mainly follow the sample-wise classification paradigm, which overlooks the relation-dependent training nature of VPR, where supervision is typically formed by image pairs rather than independent images. To address this issue, we propose a place-wise dataset pruning framework tailored for VPR. Instead of pruning individual images, our method treats each place as the basic pruning unit and introduces two complementary novel metrics, i.e., intra-place diversity (IPD) and inter-place similarity (IPS), to evaluate the training value of each place. By jointly considering these two metrics, our method ranks all places and constructs a compact yet informative coreset, thereby allowing the pruned dataset to still support the training of robust and discriminative VPR models. Extensive experiments demonstrate that our method consistently outperforms state-of-the-art DP baselines under different pruning ratios while reducing selection and training costs. Moreover, by pruning a merged dataset roughly 3.5$\times$ the size of GSV-Cities to a comparable scale, our coreset maintains highly competitive performance, achieving 94.5\% R@1 on MSLS-val and 97.0\% R@1 on Nordland with only NetVLAD. Codes will be made publicly available.
