kaanakan/stretchbev
Official implementation of our ECCV paper "StretchBEV: Stretching Future Instance Prediction Spatially and Temporally"
This project helps self-driving car engineers predict the future movement and location of other vehicles, pedestrians, and dynamic objects around an autonomous vehicle. It takes bird's-eye view sensor data as input and outputs a probabilistic forecast of where each object will be in the coming seconds, covering both nearby and distant areas. This is for engineers designing and testing perception and prediction systems for autonomous driving.
No commits in the last 6 months.
Use this if you need to generate diverse and accurate future predictions of dynamic objects for autonomous driving systems, especially for longer time horizons and larger spatial areas.
Not ideal if you are looking for a system to control the vehicle's actions or if your primary need is real-time obstacle detection without future prediction capabilities.
Stars
44
Forks
2
Language
Python
License
MIT
Category
Last pushed
Oct 06, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/kaanakan/stretchbev"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.