AI Paper Atlas

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

Knowledge-Guided Multi-Label Few-Shot Learning for General Image Recognition

Tianshui Chen, Liang Lin, Riquan Chen, Xiaolu Hui, Hefeng Wu

TPAMI 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 197 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 graph feature propagation module (ablation)KGGR · MS-COCO multi-label recognitionmAP: 80.6 -> 84.3Improved
add semantically guided attention (ablation)KGGR · MS-COCO multi-label recognitionmAP: 80.9 -> 84.3Improved
add knowledge-embedded feature propagation (ablation)KGGR · MS-COCO multi-label recognitionmAP: 82.2 -> 84.3Improved
add graph semantic propagation (ablation)KGGR · MS-COCO multi-label recognitionmAP: 83.8 -> 84.3Improved
add the full KGGR framework to ResNet-101 (ablation)ResNet-101 · MS-COCO multi-label recognitionmAP: 80.3 -> 84.3Improved
use cross-entropy instead of Euclidean loss (ablation)KGGR · MS-COCO multi-label recognitionmAP: 79.2 -> 84.3Improved
add graph feature propagation module (ablation)KGGR · MS-COCO few-shot (1-shot)mAP: 44.1 -> 52.3Improved
add graph feature propagation module (ablation)KGGR · MS-COCO few-shot (5-shot)mAP: 56.0 -> 63.5Improved
add graph semantic propagation (ablation)KGGR · MS-COCO few-shot (1-shot)mAP: 50.4 -> 52.3Improved
add graph semantic propagation (ablation)KGGR · MS-COCO few-shot (5-shot)mAP: 61.0 -> 63.5Improved
add COCO pretrainingKGGR · PASCAL VOC 2007mAP: 93.6 -> 95.0Improved
replace GoogleNet-v3 backbone with ResNet-101KGGR · MS-COCO few-shot (1-shot)mAP: 49.4 -> 52.3Improved

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