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 ∼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.</p>\n","updatedAt":"2026-07-22T04:21:57.226Z","author":{"_id":"679d9ac8fb15b4e60a79d39e","avatarUrl":"/avatars/0135ab80973079e4e5ad7ccafba7ca79.svg","fullname":"Tianyi","name":"research-WM","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8628426194190979},"editors":["research-WM"],"editorAvatarUrls":["/avatars/0135ab80973079e4e5ad7ccafba7ca79.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.15491","authors":[{"_id":"6a5ea3594fe5d1d13e84aa75","user":{"_id":"679d9ac8fb15b4e60a79d39e","avatarUrl":"/avatars/0135ab80973079e4e5ad7ccafba7ca79.svg","isPro":false,"fullname":"Tianyi","user":"research-WM","type":"user","name":"research-WM"},"name":"Tianyi Gao","status":"claimed_verified","statusLastChangedAt":"2026-07-21T09:34:09.666Z","hidden":false},{"_id":"6a5ea3594fe5d1d13e84aa76","name":"Jiayu Lin","hidden":false},{"_id":"6a5ea3594fe5d1d13e84aa77","name":"Danielle Beaulieu","hidden":false},{"_id":"6a5ea3594fe5d1d13e84aa78","name":"Nathan Jacobs","hidden":false}],"publishedAt":"2026-07-16T00:00:00.000Z","submittedOnDailyAt":"2026-07-22T00:00:00.000Z","title":"Trajectory-aware Cross-view Geo-localization with Sequential Observations","submittedOnDailyBy":{"_id":"679d9ac8fb15b4e60a79d39e","avatarUrl":"/avatars/0135ab80973079e4e5ad7ccafba7ca79.svg","isPro":false,"fullname":"Tianyi","user":"research-WM","type":"user","name":"research-WM"},"summary":"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 sim39K 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/.","upvotes":1,"discussionId":"6a5ea35a4fe5d1d13e84aa79","projectPage":"https://humblegamer.github.io/trajloc/","githubRepo":"https://github.com/mvrl/TrajLoc","githubRepoAddedBy":"user","githubStars":0,"organization":{"_id":"652fef7c8f61edf213bbbab2","name":"MVRL","fullname":"Multimodal Vision Research Laboratory @ WashU","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/652ffca9729ec1a37e4e7915/P94G4gdINp2-T2IImrUg5.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"664d930f4b870dd167473c1c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/664d930f4b870dd167473c1c/TXVEPGvkhftdI_xE1mluu.jpeg","isPro":false,"fullname":"Andy Guan","user":"andytonglove","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"652fef7c8f61edf213bbbab2","name":"MVRL","fullname":"Multimodal Vision Research Laboratory @ WashU","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/652ffca9729ec1a37e4e7915/P94G4gdINp2-T2IImrUg5.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.15491.md","query":{}}">
Trajectory-aware Cross-view Geo-localization with Sequential Observations
Published on Jul 16
· Submitted by Tianyi on Jul 22 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 sim39K 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/.
Community
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 ∼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.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2607.15491 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.15491 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2607.15491 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.