AI Paper Atlas

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

RandomRooms: Unsupervised Pre-Training From Synthetic Shapes and Randomized Layouts for 3D Object Detection

Yongming Rao, Benlin Liu, Yi Wei, Jiwen Lu, Cho-Jui Hsieh, Jie Zhou

ICCV 2021 · Detection and segmentation › 3D detection and driving · approach: Unified LiDAR 3D detection · 49 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 RandomRooms pre-trainingVoteNet · ScanNetV2 valmAP@0.25: 58.6 -> 61.3Improved
add RandomRooms pre-trainingH3DNet · ScanNetV2 valmAP@0.5: 48.1 -> 51.5Improved
add RandomRooms pre-trainingVoteNet · SUN RGB-D valmAP@0.5: 32.9 -> 35.4Improved
add RandomRooms pre-trainingH3DNet · SUN RGB-D valmAP@0.25: 60.1 -> 61.6Improved
reduce training data to 5% with RandomRooms pre-trainingVoteNet · ScanNetV2 val (5% data)mAP@0.25: 12.6 -> 19.1Improved
reduce training data to 5% with RandomRooms pre-trainingH3DNet · ScanNetV2 val (5% data)mAP@0.25: 26.6 -> 32.0Improved
pre-train on ShapeNet instead of ScanNetV2 (ablation)VoteNet RandomRooms · ScanNetV2 valmAP@0.25: 60.2 -> 61.3Improved
pre-train on both ShapeNet and ScanNetV2 (ablation)VoteNet RandomRooms · ScanNetV2 valmAP@0.25: 61.3 -> 61.5Improved
use instance-level instead of point-level contrastive loss (ablation)VoteNet pre-training · ScanNetV2 valmAP@0.25: 59.2 -> 61.3Improved
add segmentation loss to instance-level contrastive (ablation)VoteNet pre-training · ScanNetV2 valmAP@0.25: 61.3 -> 61.5No clear effect
add gravity principle in layout generation (ablation)RandomRooms · ScanNetV2 valmAP@0.25: 60.5 -> 61.3Improved
add floor and wall to generated scene (ablation)RandomRooms · ScanNetV2 valmAP@0.25: 60.7 -> 61.3Improved

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