vita-epfl/s-attack

[CVPR 2025] Official implementation of three papers "Certified Human Trajectory Prediction", "Vehicle trajectory prediction works, but not everywhere", and "Are socially-aware trajectory prediction models really socially-aware?".

48
/ 100
Emerging

This helps evaluate and improve models that predict how people or vehicles will move. It takes an existing trajectory prediction model and real-world movement data, then outputs insights into the model's reliability in different social or environmental situations. Autonomous vehicle engineers, robotics researchers, and urban planners can use this to build safer and more dependable predictive systems.

120 stars.

Use this if you need to rigorously test and understand the limitations of your trajectory prediction models, especially regarding their ability to handle social interactions or diverse environmental conditions.

Not ideal if you are looking for a pre-built trajectory prediction model, rather than tools to analyze and certify an existing one.

autonomous-driving robotics-navigation pedestrian-dynamics model-robustness predictive-modeling
No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

120

Forks

17

Language

Python

License

AGPL-3.0

Last pushed

Dec 31, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/vita-epfl/s-attack"

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