Publications de Li Mi | EPFL Graph Search

Unsupervised Multi-view UAV Image Geo-localization via Iterative Rendering

Unmanned Aerial Vehicle (UAV) Cross-View Geo-Localization (CVGL) poses significant challenges due to the substantial view discrepancies between oblique UAV images and overhead satellite images. Existing methods heavily rely on supervised learning with labe ...

2025

Knowledge-Aware Text–Image Retrieval for Remote Sensing Images

Devis Tuia, Li Mi, Javiera Francisca Castillo Navarro

Image-based retrieval in large Earth observation archives is challenging because one needs to navigate across thousands of candidate matches only with the query image as a guide. By using text as information supporting the visual query, the retrieval syste ...

Institute of Electrical and Electronics Engineers (IEEE)2024

ConGeo: Robust Cross-View Geo-Localization Across Ground View Variations

Devis Tuia, Antoine Bosselut, Li Mi, Javiera Francisca Castillo Navarro, Syrielle Mathilde Marie Montariol

Cross-view geo-localization aims at localizing a ground-level query image by matching it to its corresponding geo-referenced aerial view. In real-world scenarios, the task requires accommodating diverse ground images captured by users with varying orientat ...

Springer Cham2024

ConVQG: Contrastive Visual Question Generation with Multimodal Guidance

Devis Tuia, Antoine Bosselut, Li Mi, Javiera Francisca Castillo Navarro, Syrielle Mathilde Marie Montariol

Asking questions about visual environments is a crucial way for intelligent agents to understand rich multi-faceted scenes, raising the importance of Visual Question Generation (VQG) systems. Apart from being grounded to the image, existing VQG systems can ...

Association for the Advancement of Artificial Intelligence (AAAI)2024

Knowledge-Aware Visual Question Generation for Remote Sensing Images

Devis Tuia, Li Mi, Javiera Francisca Castillo Navarro

With the rapid development of remote sensing image archives, asking questions about images has become an effective way of gathering specific information or performing image retrieval. However, automatically generated image-based questions tend to be simpli ...

IEEE2024

Civil war hinders crop production and threatens food security in Syria

Xi Li, Li Mi

Assessing the impact of violent conflict on Syrian agriculture is challenging given data limitations and attributability issues. Using satellite data at 30 m spatial resolution, we found that the extent of productive cropland showed greater interannual var ...

NATURE PORTFOLIO2022

Knowledge-Aware Cross-Modal Text-Image Retrieval for Remote Sensing Images

Devis Tuia, Christel Marie Tartini-Chappuis, Siran Li, Li Mi

Image-based retrieval in large Earth observation archives is difficult, because one needs to navigate across thousands of candidate matches only with the proposition image as a guide. By using text as a query language, the retrieval system gains in usabili ...

2022