AI Paper Atlas

Research map / Point clouds / Point cloud classification

SimIPU: Simple 2D Image and 3D Point Cloud Unsupervised Pre-training for Spatial-Aware Visual Representations

Zhenyu Li, Zehui Chen, Ang Li, Liangji Fang, Qinhong Jiang, Xianming Liu, Junjun Jiang, Bolei Zhou, Hang Zhao

AAAI 2022 · Point clouds › Point cloud classification · approach: Point-language and open-vocabulary 3D · 53 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
Pre-train with SimIPU on KITTI image–point-cloud pairsMVXNet with Moca · KITTI validationOverall AP3D, moderate (%): 68.58 -> 70.88Improved
Pre-train with SimIPU on Waymo image–point-cloud pairsMVXNet · Waymo validationOverall mAP, L1/L2 (%): 65.57/61.53 -> 66.92/63.01Improved
Pre-train with SimIPU on Waymo image–point-cloud pairsResNet-50 depth baseline · KITTI Eigen splitRMS: 3.575 -> 2.840Improved
Initialize SimIPU pre-training from Super-IN, then double fine-tuneResNet-50 depth baseline · KITTI Eigen splitRMS: 2.712 -> 2.592Improved
Initialize SimIPU pre-training from Super-IN, then double fine-tuneFCOS3D · Nuscenes validationAP (%): +0.65Improved
Add intra-modal point-cloud contrastive learning (ablation)SimIPU · KITTI validationOverall AP3D, moderate (%): 69.19 -> 70.88Improved
Add inter-modal feature interaction (ablation)SimIPU · KITTI validationOverall AP3D, moderate (%): 68.58 -> 70.88Improved
Replace Hungarian matching with greedy assignment (ablation)SimIPU · KITTI validationOverall AP3D, moderate (%): 70.88 -> 68.20Hurt
Increase the amount of Waymo pre-training data (ablation)SimIPU · KITTI monocular depth estimation and Nuscenes monocular 3D detectionRMSE and APImproved
Double fine-tune from other unsupervised pre-training (MoCo-v2) instead of SimIPUSingle-modality downstream models · Monocular depth estimation and monocular 3D object detectionConvergenceHurt

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