ykotseruba/PedestrianActionBenchmark

Code and models for the WACV 2021 paper "Benchmark for evaluating pedestrian action prediction"

43
/ 100
Emerging

This project helps researchers and engineers in autonomous vehicle development to evaluate and improve systems that predict when pedestrians will cross the road. It takes video data of pedestrians and outputs predictions of their crossing behavior, allowing users to benchmark different prediction algorithms. This is for professionals working on self-driving cars or advanced driver-assistance systems.

No commits in the last 6 months.

Use this if you need to rigorously compare and test different algorithms for predicting pedestrian crossing actions in real-world scenarios for autonomous systems.

Not ideal if you are looking for a ready-to-use pedestrian detection or general object tracking solution without the need for predictive action benchmarking.

autonomous-driving pedestrian-behavior-prediction ADAS computer-vision robotics
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

63

Forks

19

Language

Python

License

MIT

Last pushed

May 12, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ykotseruba/PedestrianActionBenchmark"

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