search

ika-rwth-aachen / MultiCorrupt

MultiCorrupt: A benchmark for robust multi-modal 3D object detection, evaluating LiDAR-Camera fusion models in autonomous driving. Includes diverse corruption types (e.g., misalignment, miscalibration, weather) and severity levels. Assess model performance under challenging conditions.
MIT License
41 stars 5 forks source link

Support for other datasets #1

Closed Barath19 closed 7 months ago

Barath19 commented 7 months ago

Hello team,

Do you plan to support other dataset like kitti and waymo?

TillBeemelmanns commented 7 months ago

Hi,

It would be interesting to support other datasets. However, most published multi-modal models are designed and tuned for nuScenes. Only a few evaluate their models on Waymo. Kitti seems to be a bit outdated.

Given the limited research value, adapting the script to other datasets is currently not a priority. However, I would not completely exclude that idea and I am also open for PRs.

Bests, Till

Barath19 commented 7 months ago

Yeah thanks for the enlightenment!! If needed I will open a PR.