AI Paper Atlas

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

Object Detection with a Unified Label Space from Multiple Datasets

Xiangyun Zhao, Samuel Schulter, Gaurav Sharma, Yi-Hsuan Tsai, Manmohan Chandraker, Ying Wu

ECCV 2020 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Pseudo-labeling for semi-supervised detection · 68 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 pseudo labeling instead of UOD+Mergeunified detector · Ablation setup, MIXAP50: 46.1 to 52.2Improved
add pseudo labeling instead of ignoring label space ambiguity (ablation)unified detector · Ablation setup, V-on-CAP50: 36.5 -> 62.2Improved
add partial-annotation sum loss instead of ignoring ambiguity (ablation)unified detector · Ablation setup, MIXAP50: 43.7 -> 44.1No clear effect
add minimum-entropy regularization to partial-loss sum (ablation)unified detector · Ablation setup, MIXAP50: 44.1 -> 43.9No clear effect
replace sum with max in partial-annotation loss (ablation)unified detector · Ablation setup, MIXAP50: 44.1 -> 44.6No clear effect
use hard-thresholding instead of soft-thresholding in the pseudo-label weighting (ablation)pseudo labeling · Ablation setup, MIXAP50: 51.6 -> 52.2Improved
add regression loss on pseudo ground truth (ablation)pseudo labeling · Ablation setup, MIXAP50: 52.0 -> 52.2Improved
use shared-backbone UOD+Merge instead of individual detectors (ablation)UOD · Ablation setup, MIXAP50: 46.9 -> 46.1No clear effect
threshold pseudo detections to confident ones only (ablation)pseudo labeling of VOC detector on COCO train · COCO train pseudo labelsprecision / recall: precision 0.29 -> 0.79, recall 0.7 -> 0.4Mixed
add pseudo labeling instead of UOD+Mergeunified detector · Setting AAP50: 59.3 -> 61.1Improved
add pseudo labeling instead of UOD+Mergeunified detector · Setting BAP50: 43.7 -> 48.5Improved
add pseudo labeling instead of partial-annotation lossunified detector · Setting BAP50: 41.2 -> 48.5Improved

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