Trajectory-aware Cross-view Geo-localization with Sequential Observations
Computer Vision and Pattern Recognition
Summary
The authors study how to find locations by matching ground-level videos or route descriptions to satellite images. They introduce a new dataset called SeqGeo-VL that includes videos, text routes, and satellite images for the same places. They also propose TrajLoc, a system that combines visual video data and text descriptions to improve location matching. Additionally, they create a module called TrajMod to make the system aware of the route's shape, helping it understand spatial information better. Experiments show their approach works better than previous methods.
Authors
Tianyi Gao, Jiayu Lin, Danielle Beaulieu, Nathan Jacobs
Abstract
Cross-view geo-localization matches ground-level observations against geo-tagged satellite imagery. Recent methods show that sequential queries such as video clips yield richer spatiotemporal cues than single images, yet they overlook a complementary sequential modality: route descriptions -- which capture the same trajectory at a higher level of abstraction and are often the only input available (e.g., a user directing an autonomous vehicle to a pickup point). To bridge this gap, we introduce SeqGeo-VL, a dataset of $\sim$39K video-text-satellite triplets, and TrajLoc, a unified framework capable of processing both video clips and route descriptions. By leveraging both dense visual and abstract linguistic semantics, TrajLoc enables these modalities to mutually reinforce cross-view matching. We further propose TrajMod, a lightweight module that conditions query embeddings on trajectory geometry, yielding spatially-aware representations. Experiments show that TrajLoc achieves substantial gains over state-of-the-art methods on both video and text geo-localization. The project page is available at https://humblegamer.github.io/trajloc/.