Research map / Label-efficient and robust learning / Recognition and long-tail
Attention-Driven Dynamic Graph Convolutional Network for Multi-Label Image Recognition
Jin Ye, Junjun He, Xiaojiang Peng, Wenhao Wu, Yu Qiao
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add SAM and D-GCN to ResNet-101 (full ADD-GCN) (ablation) | ResNet-101 · MS-COCO | mAP: 79.7 -> 85.2 | Improved |
| add SAM and D-GCN to ResNet-101 (full ADD-GCN) (ablation) | ResNet-101 · Pascal VOC 2007 | mAP: 90.8 -> 96.0 | Improved |
| use dynamic graph instead of static graph (ablation) | D-GCN · MS-COCO | mAP: 82.9 -> 83.7 | Improved |
| use dynamic graph instead of static graph (ablation) | D-GCN · Pascal VOC 2007 | mAP: 94.5 -> 94.9 | Improved |
| stack static then dynamic graph (S->D) instead of dynamic then static (ablation) | D-GCN · MS-COCO | mAP: 84.5 -> 85.2 | Improved |
| stack static then dynamic graph (S->D) instead of dynamic then static (ablation) | D-GCN · Pascal VOC 2007 | mAP: 95.0 -> 96.0 | Improved |
| stack static then dynamic graph (S->D) instead of parallel add fusion (ablation) | D-GCN · MS-COCO | mAP: 84.0 -> 85.2 | Improved |
| use per-category binary classifiers (Bi) instead of max aggregation (ablation) | ADD-GCN · MS-COCO | mAP: 83.9 -> 85.2 | Improved |
| use per-category binary classifiers (Bi) instead of sum aggregation (ablation) | ADD-GCN · Pascal VOC 2007 | mAP: 94.7 -> 96.0 | Improved |
| classify before global max pooling (cls->GMP) instead of GMP->cls (ablation) | SAM · MS-COCO | mAP: 84.1 -> 85.2 | Improved |
| classify before global max pooling (cls->GMP) instead of GAP->cls (ablation) | SAM · Pascal VOC 2007 | mAP: 94.8 -> 96.0 | Improved |
| classify before global max pooling (cls->GMP) instead of GAP->cls (ablation) | SAM · MS-COCO | mAP: 85.0 -> 85.2 | Improved |
Closest papers by meaning
- Multi-Label Image Recognition With Graph Convolutional Networks (CVPR 2019 · similarity 0.93)
- Learning Graph Convolutional Networks for Multi-Label Recognition and Applications (TPAMI 2021 · similarity 0.93)
- GKGNet: Group K-Nearest Neighbor based Graph Convolutional Network for Multi-Label Image Recognition (ECCV 2024 · similarity 0.91)
- Multi-Label Classification with Label Graph Superimposing (AAAI 2020 · similarity 0.90)
- GLDL: Graph Label Distribution Learning (AAAI 2024 · similarity 0.87)
- Enhancing the Influence of Labels on Unlabeled Nodes in Graph Convolutional Networks (ICML 2025 · similarity 0.87)
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.
