Research map / Image generation / Editing and style transfer
Dependency-aware Attention Control for Unconstrained Face Recognition with Image Sets
Xiaofeng Liu, B.V.K Vijaya Kumar, Chao Yang, Qingming Tang, Jane You
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 dependency-aware actor-critic attention (DAC on) instead of independent quality weighting | NAN baseline · IJB-A | TAR@FAR=0.01: 0.941 -> 0.951 | Improved |
| add dependency-aware actor-critic attention (DAC on) | NAN baseline · IJB-A | Rank-1 accuracy: 0.958 -> 0.970 | Improved |
| add trust-region experience replay (off-policy) (ablation) | DAC(on) · IJB-A | Rank-1 accuracy: 0.970 -> 0.972 | Improved |
| add trust-region experience replay (off-policy) (ablation) | DAC(on) · IJB-A | TAR@FAR=0.1: 0.980 -> 0.981 | Improved |
| add pose-guided representation (PGR) (ablation) | DAC(off) · IJB-A | TAR@FAR=0.01: 0.953 -> 0.954 | Improved |
| add pose-guided representation (PGR) (ablation) | DAC(off) · IJB-A | Rank-1 accuracy: 0.972 -> 0.973 | Improved |
| add pose-guided representation (PGR) (ablation) | DAC(off) · IJB-A | TAR@FAR=0.1: 0.981 -> 0.981 | No clear effect |
| add dependency-aware actor-critic attention (DAC on) | NAN baseline · YTF | verification accuracy: 0.9572 -> 0.9597 | Improved |
| add trust-region experience replay (off-policy) (ablation) | DAC(on) · YTF | verification accuracy: 0.9597 -> 0.9601 | Improved |
| add dependency-aware actor-critic attention (DAC on) | NAN baseline · Celebrity-1000 closed-set (1000 subjects) | rank-1 accuracy: 0.7717 -> 0.8278 | Improved |
| add trust-region experience replay (off-policy) (ablation) | DAC(on) · Celebrity-1000 closed-set (1000 subjects) | rank-1 accuracy: 0.8278 -> 0.8353 | Improved |
| add dependency-aware actor-critic attention (DAC on) | NAN baseline · Celebrity-1000 open-set (800 subjects) | rank-1 accuracy: 0.7987 -> 0.8205 | Improved |
Closest papers by meaning
- Attention-Aware Deep Reinforcement Learning for Video Face Recognition (ICCV 2017 · similarity 0.88)
- Permutation-Invariant Feature Restructuring for Correlation-Aware Image Set-Based Recognition (ICCV 2019 · similarity 0.85)
- FaceCoresetNet: Differentiable Coresets for Face Set Recognition (AAAI 2024 · similarity 0.84)
- Self-Critical Attention Learning for Person Re-Identification (ICCV 2019 · similarity 0.83)
- Quality Aware Network for Set to Set Recognition (CVPR 2017 · similarity 0.83)
- Deep Reinforced Attention Learning for Quality-Aware Visual Recognition (ECCV 2020 · similarity 0.83)
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.
