Locate where in the world an image was taken with TransLocator
Locate where in the world an image was taken with TransLocator
Where in the World is this Image? Transformer-based Geo-localization in the Wild
arXiv paper abstract https://arxiv.org/abs/2204.13861v1
arXiv PDF paper https://arxiv.org/pdf/2204.13861v1.pdf
Predicting the geographic location (geo-localization) from a single ground-level RGB image taken anywhere in the world is a very challenging problem.
... challenges ... due to different environmental scenarios, drastic variation in the appearance ... depending on the time of the day, weather, season, and ... prediction ... from ... image ... having only a few geo-locating cues.
For these reasons, most existing works are restricted to specific cities, imagery, or worldwide landmarks.
... propose TransLocator, a unified dual-branch transformer network that attends to tiny details over the entire image and produces robust feature representation under extreme appearance variations.
... takes an RGB image and its semantic segmentation map as inputs, interacts between its two parallel branches after each transformer layer, and simultaneously performs geo-localization and scene recognition in a multi-task fashion.
... obtain ... continent-level accuracy improvement over the state-of-the-art. ... on real-world test ... more effective than previous methods.
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments