AI Paper Atlas

Research map / 3D vision / 3D generation

OccWorld: Learning a 3D Occupancy World Model for Autonomous Driving

Wenzhao Zheng, Weiliang Chen, Yuanhui Huang, Borui Zhang, Yueqi Duan, Jiwen Lu

ECCV 2024 · 3D vision › 3D generation · approach: Vision-based 3D occupancy prediction · 156 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 spatial aggregation before temporal attention (ablation)OccWorld-O · Occ3D-nuScenes forecastingAvg. mIoU (%): 10.07 -> 17.14Improved
use temporal attention instead of simple convolution (ablation)OccWorld-O · Occ3D-nuScenes forecastingAvg. mIoU (%): 8.98 -> 17.14Improved
add ego token (joint modeling of scene and ego) (ablation)OccWorld-O · Occ3D-nuScenes forecastingAvg. mIoU (%): 15.13 -> 17.14Improved
use temporal attention for the ego token instead of MLP (ablation)OccWorld-O · nuScenes planningAvg. L2 (m): 5.89 -> 1.17Improved
use codebook size 1024 instead of 512 (ablation)scene tokenizer · Occ3D-nuScenes forecastingAvg. mIoU (%): 17.14 -> 16.30Hurt
increase latent spatial resolution of tokenizer (ablation)scene tokenizer · Occ3D-nuScenes forecastingAvg. forecasting mIoU (%): reconstruction mIoU 66.38 -> 78.12; forecasting 17.14 -> 12.38Mixed
reduce tokenizer channel dimension to 64 (ablation)scene tokenizer · Occ3D-nuScenes forecastingAvg. mIoU (%): 17.14 -> 14.67Hurt
use ground-truth 3D occupancy instead of predicted camera occupancy as inputOccWorld · Occ3D-nuScenes forecastingAvg. mIoU (%): 8.62 -> 17.14Improved
use world-model paradigm with GT occupancy instead of OccNet pipelineplanning · nuScenes planningAvg. L2 (m): 2.25 -> 1.17Improved
OccWorld-O versus UniAD at long horizon planningOccWorld-O · nuScenes planningL2 at 3s (m): 1.99 vs 1.65 (UniAD)Hurt

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