用轻量查询网络+地理记忆库，让手机也能高效认路。
Towards Test-time Efficient Visual Place Recognition via Asymmetric Query Processing
- 异构网络搭配：大模型离线提特征，小模型在线快速匹配
- 引入地理记忆库，省去耗时的k-NN计算，提速40%以上
- 适合移动端、无人机等资源受限场景的视觉定位
视觉位置识别（VPR）借助DINOv2等大容量基础模型取得了显著进展，但其高昂的计算开销使其难以在资源受限设备上部署。本文提出一种高效的非对称VPR框架：使用高容量画廊模型离线提取特征，搭配轻量级查询网络实现在线处理。核心挑战在于异构网络间的兼容性，传统方法依赖计算成本高的k-NN兼容训练。为此，我们提出基于地理信息元数据的地理记忆库，有效组织画廊特征，无需进行全量k-NN计算。此外，引入隐式嵌入增强技术，提升轻量查询网络对特征变化的建模能力。大量实验表明，该方法显著降低计算开销，且优于现有异构检索技术，在资源受限环境中展现新范式。代码已开源。
原文摘要 · Abstract (English)
Visual Place Recognition (VPR) has advanced significantly with high-capacity foundation models like DINOv2, achieving remarkable performance. Nonetheless, their substantial computational cost makes deployment on resource-constrained devices impractical. In this paper, we introduce an efficient asymmetric VPR framework that incorporates a high-capacity gallery model for offline feature extraction with a lightweight query network for online processing. A key challenge in this setting is ensuring compatibility between these heterogeneous networks, which conventional approaches address through computationally expensive k-NN-based compatible training. To overcome this, we propose a geographical memory bank that structures gallery features using geolocation metadata inherent in VPR databases, eliminating the need for exhaustive k-NN computations. Additionally, we introduce an implicit embedding augmentation technique that enhances the query network to model feature variations despite its limited capacity. Extensive experiments demonstrate that our method not only significantly reduces computational costs but also outperforms existing asymmetric retrieval techniques, establishing a new aspect for VPR in resource-limited environments. The code is available at https://github.com/jaeyoon1603/AsymVPR
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。