AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

UniM2AE: Multi-modal Masked Autoencoders with Unified 3D Representation for 3D Perception in Autonomous Driving

Jian Zou, Tianyu Huang, Guanglei Yang, Zhenhua Guo, Tao Luo, Chun-Mei Feng, Wangmeng Zuo

ECCV 2024 · Detection and segmentation › 3D detection and driving · approach: LiDAR perception for driving · 18 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
interact camera and LiDAR features in 3D volume space instead of BEV during pre-training (ablation)UniM2AE · nuScenes val detectionNDS: 64.3 -> 65.2Improved
add multi-modal interaction (MMIM) in unified space versus merging two uni-modal pre-trainings (ablation)UniM2AE · nuScenes val detectionNDS: 63.1 -> 65.2Improved
pre-train with UniM2AE versus random initializationfusion detector · nuScenes val, 20% labels (C+L)mAP: 51.5 -> 55.9Improved
pre-train with PiMAE instead of random init (project points to image plane)fusion detector · nuScenes val, 20% labels (C+L)mAP: 51.5 -> 52.5No clear effect
transfer pre-trained MMIM when fine-tuning BEVFusion-SST (ablation)BEVFusion-SST · nuScenes val BEV map segmentationmIoU: 61.4 -> 67.8Improved
raise z-axis layers beyond two (ablation)UniM2AE · nuScenes val detectionNDS: 58.3 -> 58.4No clear effect
use one z-layer instead of two (ablation)UniM2AE · nuScenes val detectionNDS: 58.3 -> 56.6Hurt
raise masking ratio to 80% for both modalities (ablation)UniM2AE · nuScenes val detectionmAP: 64.5 -> 63.3Hurt
add UniM2AE pre-trained weights without MMIM transfer to LiDAR-only TransFusion-L-SSTTransFusion-L-SST · nuScenes val 3D detectionNDS: 69.9 -> 70.4Improved
add UniM2AE pre-trained encoders without MMIM to BEVFusion-SSTBEVFusion-SST · nuScenes val BEV map segmentationmIoU: 61.3 -> 61.4No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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