ViTAE-Transformer/SimDistill

The official repo for [AAAI 2024] "SimDistill: Simulated Multi-modal Distillation for BEV 3D Object Detection""

31
/ 100
Emerging

This project helps self-driving car engineers improve how accurately their vehicles detect 3D objects like cars and pedestrians using only camera images. It takes raw camera video feeds and outputs a precise Bird's-Eye-View (BEV) map showing detected objects and their 3D positions, without needing expensive LiDAR sensors. Autonomous vehicle perception system developers would use this to enhance camera-only detection systems.

No commits in the last 6 months.

Use this if you need to significantly boost the 3D object detection accuracy of camera-based autonomous driving systems to approach the performance of systems using both LiDAR and cameras, but with the cost-effectiveness of cameras alone.

Not ideal if your autonomous driving system already heavily relies on LiDAR and you are not looking to improve camera-only detection performance or reduce sensor costs.

autonomous-driving 3d-object-detection perception-systems sensor-fusion camera-vision
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

40

Forks

3

Language

Python

License

Apache-2.0

Last pushed

May 16, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ViTAE-Transformer/SimDistill"

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