Pinpoint: Grounded Worldwide Image Geolocation via Cross-Source Retrieval and Reranking
2026-06-02 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors designed a system called Pinpoint to figure out where photos were taken by looking at the images. They combined two types of pictures: regular photos from the internet and detailed street-view images, using the strengths of both. Their method first finds possible locations using a shared matching system and then picks the best spot by paying attention to nearby clues. This approach is faster and easier to repeat than others because it doesn't use big language models. Pinpoint performed better than previous methods on common tests involving both internet and street-view photos.
image geolocationstreet-view imagerycontrastive learningimage-GPS embeddingretrieve-and-rerankattention-based rerankermultimodal modelsgeospatial retrievalbenchmark datasets
Authors
Nika Chuzhoy, Brian Hu, Amit A. Arora, Jae Ro, Sarthak S. Sahu
Abstract
Image geolocation aims to estimate where a photograph was taken from its visual content. At worldwide scale, this remains challenging because visual evidence is often ambiguous, diverse, and unevenly distributed. Prior work has typically treated geolocation of ordinary internet photos and street-view imagery as separate tasks, despite their complementary strengths: internet photos better match the appearance distribution of user-captured queries, while street-view imagery provides denser, geographically grounded coverage. We present Pinpoint, a retrieve-and-rerank architecture that combines both sources in a coarse-to-fine pipeline. A contrastive image-GPS embedder is trained on both user-uploaded Flickr photos and street-view imagery, learning a shared image-GPS embedding space that is used to retrieve candidate locations. An attention-based reranker then rescores retrieved candidates by combining candidate-level visual and GPS features with cross-source evidence from nearby locations to ground the prediction. Unlike recent prior work, Pinpoint does not rely on multimodal large-language models, making inference faster and more reproducible. Pinpoint achieves state-of-the-art results across all metrics on standard benchmarks for internet photos (IM2GPS3k and YFCC4k) and street-view imagery (OSV-5M).