MultimodalGeo/GeoText-1652

An offical repo for ECCV 2024 Towards Natural Language-Guided Drones: GeoText-1652 Benchmark with Spatial Relation Matching

26
/ 100
Experimental

This project provides a unique dataset and model for developing natural language-guided drones. It takes in drone, satellite, or ground camera images and natural language descriptions, and outputs bounding boxes linked to specific textual elements, enabling drones to understand and act on spatial commands. It's intended for researchers and engineers working on autonomous drone navigation and control systems.

114 stars. No commits in the last 6 months.

Use this if you are developing or evaluating AI models that allow drones to interpret and navigate based on natural language instructions, particularly focusing on spatial relationships in real-world imagery.

Not ideal if you are looking for a plug-and-play drone control system, as this project provides a benchmark dataset and model for research and development, not an out-of-the-box solution.

drone-navigation spatial-reasoning computer-vision robotics-control geospatial-AI
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 9 / 25

How are scores calculated?

Stars

114

Forks

7

Language

Python

License

Last pushed

Jan 26, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/MultimodalGeo/GeoText-1652"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.