AI Paper Atlas

Research map / Point clouds / Segmentation and detection

Fog Simulation on Real LiDAR Point Clouds for 3D Object Detection in Adverse Weather

Martin Hahner, Christos Sakaridis, Dengxin Dai, Luc Van Gool

ICCV 2021 · Point clouds › Segmentation and detection · approach: LiDAR voxel and pillar detectors · 233 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add our physically based fog simulation to trainingPV-RCNN · STF dense fogmAP over classes (moderate): 37.51 -> 38.64Improved
add our fog simulation to trainingPointRCNN · STF dense fogmAP over classes (moderate): 35.47 -> 38.11Improved
add our fog simulation to trainingSECOND · STF dense fogmAP over classes (moderate): 34.52 -> 36.09Improved
add our fog simulation to trainingPart-A2 · STF dense fogmAP over classes (moderate): 34.34 -> 35.59Improved
add our fog simulation to trainingPointPillars · STF dense fogmAP over classes (moderate): 27.95 -> 30.68Improved
use our physical soft-target model instead of the heuristic prior fog simulation (ablation)Part-A2 · STF dense fogmAP over classes (moderate): 33.44 -> 35.59Improved
use our physical soft-target model instead of the heuristic prior fog simulation (ablation)PointRCNN · STF dense fogmAP over classes (moderate): 36.33 -> 38.11Improved
use our simulation instead of prior simulation (ablation)PV-RCNN · STF dense fogCar 3D AP@.5IoU (moderate): 46.60 -> 47.38Improved
add strongest-last filter at test time to clear-weather model (ablation)PV-RCNN · STF dense fogmAP over classes (moderate): 37.51 -> 37.98Improved
add strongest-last filter at test time to clear-weather model (ablation)SECOND · STF dense fogmAP over classes (moderate): 34.52 -> 34.43No clear effect
add strongest-last filter at test time to clear-weather model (ablation)Part-A2 · STF dense fogmAP over classes (moderate): 34.34 -> 34.25No clear effect
add our fog simulation to trainingPointRCNN · STF dense fog Cyclist AP@.25IoU (moderate)AP: 22.99 -> 22.88Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy