Research map / Graph learning / GNN architectures
Dynamic Graph Representation for Occlusion Handling in Biometrics
Min Ren, Yunlong Wang, Zhenan Sun, Tieniu Tan
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 |
|---|---|---|---|
| replace CNN/standard iris methods with DGR | iris recognition · ND-LG4000 | EER: 1.77% -> 0.62% | Improved |
| use DGR with dynamic graph matching under 30% occlusion | iris recognition · ND-LG4000 30% occluded | EER: 8.31% -> 0.85% | Improved |
| add DGR to Light CNN-9 baseline, 30% occlusion | LCNN-9 · Occluded-LFW 30% | TPR@FTR=0.1%: 22.47% -> 71.8% | Improved |
| add DGR to Light CNN-9 baseline, 50% occlusion | LCNN-9 · Occluded-LFW 50% | TPR@FTR=0.1%: 8.47% -> 25.19% | Improved |
| mask occluded areas with precise masks before the network (ablation) | DGR and CNN backbone · ND-LG4000 50% occluded | DET performance | Hurt |
| reduce graph nodes from 32 to 8 (ablation) | DGR · ND-LG4000 | EER: 0.62% -> 0.93% | Hurt |
| increase graph nodes from 32 to 128 (ablation) | DGR · ND-LG4000 | EER: 0.62% -> 0.73% | Hurt |
Closest papers by meaning
- Multiscale Dynamic Graph Representation for Biometric Recognition With Occlusions (TPAMI 2023 · similarity 0.96)
- Robust Facial Landmark Detection via Occlusion-Adaptive Deep Networks (CVPR 2019 · similarity 0.84)
- Dynamic Graph Message Passing Networks for Visual Recognition (TPAMI 2022 · similarity 0.84)
- End2End Occluded Face Recognition by Masking Corrupted Features (TPAMI 2021 · similarity 0.83)
- High-Order Information Matters: Learning Relation and Topology for Occluded Person Re-Identification (CVPR 2020 · similarity 0.83)
- OED: Towards One-stage End-to-End Dynamic Scene Graph Generation (CVPR 2024 · similarity 0.81)
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.
