NESP 5.5: Aesthetic Ratings for 3,120 Great Barrier Reef Images from Flickr

3,120 underwater images of the Great Barrier Reef were rated for aesthetic beauty by 1,585 Australian survey participants on a 10-point scale. This dataset was created in 2020 by the Griffith Institute for Tourism Research to improve an AI system for monitoring natural scene beauty. It builds on prior work from the NESP TWQ 5.5 project and includes images sourced from Flickr and a private collection.

Use Cases

• Train image classification models for aesthetic scoring based on human ratings.
• Analyze correlations between image content (e.g., coral, fish) and perceived beauty scores.
• Benchmark AI systems for automated assessment of marine landscape aesthetics.
• Study public perception and preferences regarding underwater natural scenes.

Strengths

• Includes 3,120 rated images, providing a substantial sample for model training.
• Leverages 1,585 human survey responses, offering a robust set of aesthetic judgments.
• Images were sourced from a large pool of approximately 55,000 Flickr downloads, suggesting diversity in initial collection.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect geographic and source bias, focusing solely on the Great Barrier Reef and images from Flickr.
• Row count for the rating data is unknown, which may limit suitability assessment.

Provenance

Source

Griffith Institute for Tourism Research, NESP TWQ 5.5 project, via Australian Ocean Data Network.

Collection Method

Images downloaded from Flickr via a Python application using the Flickr API, filtered for underwater content, and rated via an online Qualtrics survey.

Time Range

Survey launched in April 2020.

Freshness

Last updated 2026-07-13 14:02:59.550477; freshness should be verified.

Geography

Great Barrier Reef, Australia.