AI Paper Atlas

Research map / Point clouds / Segmentation and detection

Data Efficient 3D Learner via Knowledge Transferred from 2D Model

Ping-Chung Yu, Cheng Sun, Min Sun

ECCV 2022 · Point clouds › Segmentation and detection · approach: Scene completion and instance segmentation · 10 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 pre-training with 2D-parser pseudo-labels (panoramic) (ablation)SparseConv · ScanNet val LA 20 pointsmIoU: 51.6 -> 58.5Improved
add pre-training with 2D-parser pseudo-labels (panoramic) (ablation)O-CNN · ScanNet val LA 200 pointsmIoU: 65.0 -> 67.8Improved
use panoramic instead of perspective pseudo-label pre-training (ablation)SparseConv · ScanNet val LA 20 pointsmIoU: 56.2 -> 58.5Improved
use panoramic instead of perspective pseudo-label pre-training (ablation)O-CNN · ScanNet val LA 20 pointsmIoU: 56.2 -> 55.6Hurt
use ground-truth labels instead of pseudo-labels for perspective pre-training (ablation)SparseConv · ScanNet val LA 20 pointsmIoU: 56.2 -> 55.4No clear effect
use ground-truth labels instead of pseudo-labels for perspective pre-training (ablation)O-CNN · ScanNet val LA 20 pointsmIoU: 56.2 -> 56.9No clear effect
add semi-supervised learning on top of pre-training (ablation)Our pre-training · ScanNet val LA 20 pointsmIoU: 55.9 -> 61.1Improved
add semi-supervised learning to model trained from scratch (ablation)Trained from scratch · ScanNet val LA 20 pointsmIoU: 54.2 -> 56.6Improved
add semi-supervised learning to pre-trained modelOur pre-training · ScanNet LR test 1%mIoU: 26.6 -> 26.3No clear effect
add semi-supervised learning to pre-trained modelOur pre-training · ScanNet LR test 5%mIoU: 46.7 -> 50.8Improved
add pre-training with semi-supervised learning (ablation)O-CNN · ScanNet val LR 5%mIoU: 44.9 -> 49.4Improved
add pre-training to object classification trainingO-CNN · ModelNet40 1%accuracy: 60.2 -> 65.5Improved

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