
Img2Loc, a novel system that redefines image geolocalization as a text generation task using cutting-edge large multi-modality models (LMMs) and retrieval-augmented generation, significantly outperforms previous state-of-the-art methods without any model training.


coremsg

img2loc-leveraging-multi-modality-foundation-models-and-retrieval-augmented-generation-for-accurate-image-geolocalization


Img2Loc: Leveraging Multi-Modality Foundation Models and Retrieval-Augmented Generation for Accurate Image Geolocalization


title_rewrite


A novel deep multi-task approach that achieves state-of-the-art performance on a wide range of image geolocalization benchmarks while being robust to distribution shifts.


predicting-image-geolocations-a-novel-multi-task-approach-for-robust-and-accurate-planet-scale-localization


Predicting Image Geolocations: A Novel Multi-Task Approach for Robust and Accurate Planet-Scale Localization
