- Today
- Holidays
- Birthdays
- Reminders
- Cities
- Atlanta
- Austin
- Baltimore
- Berwyn
- Beverly Hills
- Birmingham
- Boston
- Brooklyn
- Buffalo
- Charlotte
- Chicago
- Cincinnati
- Cleveland
- Columbus
- Dallas
- Denver
- Detroit
- Fort Worth
- Houston
- Indianapolis
- Knoxville
- Las Vegas
- Los Angeles
- Louisville
- Madison
- Memphis
- Miami
- Milwaukee
- Minneapolis
- Nashville
- New Orleans
- New York
- Omaha
- Orlando
- Philadelphia
- Phoenix
- Pittsburgh
- Portland
- Raleigh
- Richmond
- Rutherford
- Sacramento
- Salt Lake City
- San Antonio
- San Diego
- San Francisco
- San Jose
- Seattle
- Tampa
- Tucson
- Washington
AI Helps Identify Dinosaur Footprints
New neural network model can classify dinosaur tracks with 80-93% accuracy.
Jan. 27, 2026 at 8:07pm
Got story updates? Submit your updates here. ›
Researchers have developed an unsupervised neural network that can analyze the shape of dinosaur footprints and classify them with high accuracy, without human bias. The model identified eight key features of footprint variation, and the researchers have made the tool publicly available as a free app called DinoTracker that allows anyone to upload photos of footprints and get instant analysis.
Why it matters
Dinosaur footprints are much more abundant in the fossil record than complete skeletons, but they are challenging to study and identify. This new AI-powered tool provides a data-driven way to analyze footprint shapes and determine what dinosaur species they likely belong to, which can help paleontologists better understand dinosaur behavior and evolution.
The details
The research team, led by Gregor Hartmann at Helmholtz-Zentrum Berlin, built the neural network model using a dataset of around 2,000 real dinosaur footprints, which they then augmented with millions of variations. The model was able to identify eight key axes of footprint variation, including digit spread and heel position. When compared to expert classifications, the model showed 80-93% agreement.
- The research was published in the scientific journal PNAS in 2026.
The players
Gregor Hartmann
A physicist and data scientist at Helmholtz-Zentrum Berlin who led the research team.
Paige dePolo
A lecturer in vertebrate biology at the Research Centre in Evolutionary Anthropology and Palaeoecology at Liverpool John Moores University, who co-authored the article.
Steve Brusatte
The author of the book "The Rise and the Fall of the Dinosaurs", which inspired Hartmann to apply his AI methods to paleontological questions.
Julius
Hartmann's young son, who was very interested in dinosaurs and sparked the idea for the research collaboration.
DinoTracker
A free public app developed by the research team that allows anyone to upload photos of dinosaur footprints and get instant analysis of what they are most similar to.
What they’re saying
“Every fossil is a miracle. It takes the perfect combination of circumstances for a fossil to form, be preserved through millions of years, and be found and recognised by human eyes.”
— Paige dePolo, Lecturer in Vertebrate Biology, Research Centre in Evolutionary Anthropology and Palaeoecology, Liverpool John Moores University (The Conversation)
What’s next
The researchers plan to continue improving the DinoTracker app and expanding its capabilities to help more people engage with and learn about dinosaur paleontology.
The takeaway
This new AI-powered tool for analyzing dinosaur footprints demonstrates how technology can be leveraged to advance scientific understanding and make it more accessible to the public. The app empowers anyone to contribute to the study of these ancient fossils, furthering our knowledge of dinosaur behavior and evolution.
Chicago top stories
Chicago events
Mar. 17, 2026
Chicago Blackhawks vs. Minnesota WildMar. 17, 2026
BASEBALL: A Celebration of BaseballMar. 17, 2026
Pattie Gonia: SAVE HER! Feat. Sequoia and VERA!




