AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Heterogeneous Semantic Transfer for Multi-label Recognition with Partial Labels

Tianshui Chen, Tao Pu, Lingbo Liu, Yukai Shi, Zhijing Yang, Liang Lin

IJCV 2024 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 36 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 IST module to SSGRL (ablation)SSGRL · MS-COCOaverage mAP: 74.1 -> 76.4Improved
add CST module to SSGRL (ablation)SSGRL · VG-200average mAP: 39.7 -> 42.7Improved
add both IST and CST (full HST) (ablation)SSGRL · MS-COCOaverage mAP: 74.1 -> 77.9Improved
use semantic-correlation pseudo labels instead of confidence-based pseudo labels (ablation)HST · VG-200average mAP: 40.8 -> 44.8Improved
learn image-specific co-occurrence instead of statistical co-occurrence (ablation)IST · MS-COCOaverage mAP: 70.7 -> 76.4Improved
add L_ist loss (ablation)IST · VOC 2007average mAP: 89.0 -> 90.0Improved
add L_cst loss (ablation)CST · MS-COCOaverage mAP: 74.8 -> 76.5Improved
use instance-prototype similarities instead of instance-level similarities (ablation)CST · MS-COCOaverage mAP: 75.5 -> 76.5Improved
use SD instead of SA for SARL (ablation)HST · MS-COCOaverage mAP: 77.7 -> 77.9No clear effect
use dynamic threshold learning instead of fixed threshold 0.7 (ablation)HST · MS-COCOaverage mAP: 77.5 -> 77.9Improved
use dynamic threshold learning instead of fixed threshold 0.1 (ablation)HST · VG-200average mAP: 40.8 -> 44.8Improved
increase number of prototypes K from 10 to 20 (ablation)CST · MS-COCOaverage mAP: K from 1 to 10 improves, K=20 decreasesHurt

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