AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

Power of Cooperative Supervision: Multiple Teachers Framework for Advanced 3D Semi-Supervised Object Detection

Jin-Hee Lee, Jae-Keun Lee, Jeseok Kim, Kwon Soon

ICCV 2025 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Pseudo-labeling for semi-supervised detection · 2 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 multiple category-specialized teachers (MPGen) instead of a single teacher (ablation)3D SSOD baseline · LiO 15% labeledmAP: 51.7 -> 53.8Improved
add PieAug on top of MPGen (ablation)MultipleTeachers · LiO 15% labeledmAP: 53.8 -> 55.0Improved
set PieAug pie angle to 45 degrees instead of 15 degrees (ablation)PieAug · 1% labeledmAP: 53.0 -> 54.2Improved
increase PieAug pie angle from 45 to 60 degrees (ablation)PieAug · 1% labeledmAP: 54.2 -> 53.0Hurt
use PieAug instead of PolarMix (ablation)MultipleTeachers · KITTI 1% labeledmAP: 36.4 -> 64.6Improved
use PieAug instead of Shuffle Data augmentation (ablation)MultipleTeachers · KITTI 1% labeledmAP: 43.7 -> 64.6Improved
use PieAug instead of PseudoAugment (ablation)MultipleTeachers · KITTI 1% labeledmAP: 62.2 -> 64.6Improved
use PieAug instead of PseudoAugment (ablation)MultipleTeachers · LiO 1% labeledmAP: 39.0 -> 39.1No clear effect
use MultipleTeachers SSL instead of HSSDA with PV-RCNNPV-RCNN · LiO 15% labeledmAP: 49.1 -> 50.5Improved
use MultipleTeachers SSL instead of HSSDA with Re-VoxelDetRe-VoxelDet · LiO 15% labeledmAP: 58.8 -> 59.7Improved
use MultipleTeachers instead of HSSDAVoxel-RCNN · LiO 1% labeledmAP: 30.7 -> 37.8Improved
use MultipleTeachers instead of HSSDAVoxel-RCNN · LiO 15% labeledmAP: 53.7 -> 55.0Improved

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