basilevh/occlusions-4d

Revealing Occlusions with 4D Neural Fields (CVPR 2022 Oral) - Official Implementation

35
/ 100
Emerging

This project helps self-driving car engineers and robotics researchers create accurate 3D models of dynamic environments where objects might be hidden or partially blocked. It takes in video streams of 3D scenes (like those from autonomous vehicle sensors) and outputs a complete, 4D (3D over time) representation of the scene, including currently occluded objects. This allows for a more comprehensive understanding of the environment for improved perception and navigation.

No commits in the last 6 months.

Use this if you need to accurately reconstruct complex, dynamic 3D environments from video data, especially when dealing with moving objects that frequently become occluded or reappear.

Not ideal if your primary goal is static 3D reconstruction, or if you are working with environments where occlusions are not a significant challenge.

autonomous-driving robotics-perception 3d-scene-reconstruction computer-vision dynamic-environment-modeling
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

78

Forks

6

Language

Python

License

MIT

Last pushed

Jan 19, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/basilevh/occlusions-4d"

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