NocPlace: Nocturnal Visual Place Recognition via Generative and Inherited Knowledge Transfer

TL;DR

NocPlace introduces a novel approach for nighttime visual place recognition by leveraging generative models and knowledge transfer, improving robustness against lighting variations and dark conditions without additional real-time computational costs.

Contribution

The paper presents NocPlace, a new method that uses a large-scale night scene dataset and generative models to enhance VPR performance across day-night domains.

Findings

Improves Eigenplaces performance by 7.6% on Tokyo 24/7 Night.

Achieves 16.8% improvement on SVOX Night.

Demonstrates robustness against dazzling lights and darkness.

Abstract

Visual Place Recognition (VPR) is crucial in computer vision, aiming to retrieve database images similar to a query image from an extensive collection of known images. However, like many vision tasks, VPR always degrades at night due to the scarcity of nighttime images. Moreover, VPR needs to address the cross-domain problem of night-to-day rather than just the issue of a single nighttime domain. In response to these issues, we present NocPlace, which leverages generative and inherited knowledge transfer to embed resilience against dazzling lights and extreme darkness in the global descriptor. First, we establish a day-night urban scene dataset called NightCities, capturing diverse lighting variations and dark scenarios across 60 cities globally. Then, an image generation network is trained on this dataset and processes a large-scale VPR dataset, obtaining its nighttime version. Finally, VPR models are fine-tuned using descriptors inherited from themselves and night-style images, which builds explicit cross-domain contrastive relationships. Comprehensive experiments on various datasets demonstrate our contributions and the superiority of NocPlace. Without adding any real-time computing resources, NocPlace improves the performance of Eigenplaces by 7.6% on Tokyo 24/7 Night and 16.8% on SVOX Night.