AI Paper Atlas

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

Exploiting Unlabeled Data with Vision and Language Models for Object Detection

Shiyu Zhao, Zhixing Zhang, Samuel Schulter, Long Zhao, Vijay Kumar B G, Anastasis Stathopoulos, Manmohan Chandraker, Dimitris N. Metaxas

ECCV 2022 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Pseudo-labeling for semi-supervised detection · 107 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 RoI head box refinement to pseudo labels (tau=0.95) (ablation)VL-PLM · COCO-ZSNovel AP50: 26.0 -> 31.3Improved
add RoI head refinement at tau=0.05 (ablation)VL-PLM · COCO-ZSNovel AP50: 14.6 -> 19.3Improved
raise pseudo-label threshold from 0.05 to 0.95 (with RoI refinement) (ablation)VL-PLM · COCO-ZSNovel AP50: 19.3 -> 31.3Improved
raise threshold from 0.95 to 0.99 (ablation)VL-PLM · COCO-ZSNovel AP50: 31.3 -> 27.2Hurt
use RPN+RoI proposals instead of selective search (ablation)VL-PLM · COCO-ZSAP@PL: 5.7 -> 25.3Improved
use RPN+RoI instead of RoI head alone (ablation)VL-PLM · COCO-ZSAP@PL: 8.8 -> 25.3Improved
use RPN+RoI instead of RPN alone (ablation)VL-PLM · COCO-ZSAP@PL: 19.7 -> 25.3Improved
fuse teacher pseudo labels with VL-PLM labels (1% COCO) (ablation)VL-PLM SSOD · COCO SSOD (1% labeled)mAP: 13.27 -> 15.35Improved
fuse teacher pseudo labels with VL-PLM labels (10% COCO) (ablation)VL-PLM SSOD · COCO SSOD (10% labeled)mAP: 24.20 -> 27.23Improved
use VL-PLM labels without teacher labels instead of teacher labels only (10% COCO) (ablation)SSOD · COCO SSOD (10% labeled)mAP: 25.98 -> 24.20Hurt
replace STAC pseudo labels with VL-PLM (1% COCO)STAC · COCO SSOD (1% labeled)mAP: 13.97 -> 17.71Improved
add more unlabeled images (1:20 ratio vs none) at 5% labeled (ablation)Faster R-CNN · COCO SSOD (5% labeled)mAP: 17.7 -> 23.7Improved

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