LandCover.ai: Aerial Imagery for Mapping Buildings, Woodlands, Water, and Roads

A dataset for automatic mapping of buildings, woodlands, water, and roads from aerial imagery, originally published with the LandCover.ai paper. The dataset was uploaded to Hugging Face by MortenTabaka and last updated on 2023-03-26. It is associated with a GitHub project for semantic segmentation.

Use Cases

• Training semantic segmentation models based on aerial imagery for land cover classification.
• Developing automatic mapping tools for buildings, woodlands, water, and roads as described in the original paper.
• Benchmarking computer vision algorithms for geospatial image analysis.
• Creating applications for urban planning or environmental monitoring based on land cover features.

Strengths

• Dataset is associated with a published research paper, suggesting a defined academic purpose.
• The dataset is linked to a specific GitHub project demonstrating a practical application for semantic segmentation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• Last updated 2023-03-26 17:28:43; freshness should be verified.

Provenance

Source

Originally published with the 'LandCover.ai: Dataset for Automatic Mapping of Buildings, Woodlands, Water and Roads from Aerial Imagery' paper.

Collection Method

Likely gathered from aerial imagery sources for the purpose of creating a labeled dataset for semantic segmentation.

Freshness

Last updated 2023-03-26 17:28:43.