An artificial intelligence application identifies dinosaur footprints with 90% accuracy… and may have found the footprints of the planet’s first birds, dating back 200 million years
Publish Date: 2026-04-06 18:00:00
Source Domain: www.ecoticias.com
- Dinosaur Footprint Analysis: The article discusses how artificial intelligence has made significant progress in identifying and classifying dinosaur footprints using an app called DinoTracker.
- AI Training: DinoTracker was trained on 1,974 dinosaur and bird footprints, learning to sort the footprints into categories based on inherent patterns and features in the tracks.
- Key Features: The algorithm analyses eight features of dinosaur footprints such as toe spread, weight distribution, heel placement, etc., achieving classification accuracy comparable to that of experts.
- Oldest Birds Discovery: The app reclassified some ancient three-toed footprints to bird-like prints, suggesting potential birds roamed Earth around 210 million years ago, which is 60 million years earlier than previously believed.
- Ecological Insights: DinoTracker provides valuable data on how various dinosaur groups moved and behaved, which contributes to understanding ancient ecosystems and how species may respond to modern environmental changes.
- Role in Research: While DinoTracker assists in footprint analysis, it does not replace human expertise. Researchers still need to verify the age and geological context of each site for conclusive findings.
- Public Accessibility: DinoTracker is freely available for everyone, from professionals to the curious public, providing a standardized way to analyze footprints.
- Collaborative Tool: The app serves as an additional resource in paleontology, highlighting patterns that human eyes often miss and acting as a collaborative tool in research.
