Research map / Label-efficient and robust learning / Recognition and long-tail
Attribute-Enhanced Face Recognition With Neural Tensor Fusion Networks
Guosheng Hu, Yang Hua, Yang Yuan, Zhihong Zhang, Zheng Lu, Sankha S. Mukherjee, Timothy M. Hospedales, Neil M. Robertson, Yongxin Yang
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 |
|---|---|---|---|
| fuse attribute features (AttNet) with GTNN | LeanFace · Multi-PIE pose -90 deg | recognition rate (%): 72.0 -> 78.3 | Improved |
| fuse attribute features (AttNet) with GTNN | LBP · Multi-PIE pose -90 deg | recognition rate (%): 1.3 -> 16.3 | Improved |
| fuse attribute features (AttNet) with GTNN | LeanFace · Multi-PIE pose +90 deg | recognition rate (%): 61.0 -> 68.0 | Improved |
| fuse attribute features (AttNet) with GTNN | LBP · Multi-PIE illumination | recognition rate (%): 57.5 -> 80.3 | Improved |
| fuse attribute features (AttNet) with GTNN | LBP · Multi-PIE expression | recognition rate (%): 71.7 -> 77.5 | Improved |
| fuse attribute features (AttNet) with GTNN | LeanFace · Multi-PIE illumination | recognition rate (%): 100 -> 100 | No clear effect |
| fuse attribute features (AttNet) with GTNN | LeanFace · CASIA NIR-VIS 2.0 | accuracy (%): 97.27 -> 99.94 | Improved |
| fuse attribute features (AttNet) with GTNN | LeanFace · LFW | accuracy (%): 99.57 -> 99.65 | Improved |
| use GTNN instead of LDA fusion (ablation) | LeanFace + AttNet · CASIA NIR-VIS 2.0 | accuracy (%): 98.33 -> 99.94 | Improved |
| use feature concatenation instead of no fusion (ablation) | LeanFace + AttNet · CASIA NIR-VIS 2.0 | accuracy (%): 97.27 -> 97.27 | No clear effect |
| use score fusion (ablation) | LeanFace + AttNet · CASIA NIR-VIS 2.0 | accuracy (%): 97.27 -> 64.24 | Hurt |
| use CCA fusion (ablation) | LeanFace + AttNet · LFW | accuracy (%): 99.57 -> 99.23 | Hurt |
Closest papers by meaning
- Attributes for Improved Attributes: A Multi-Task Network Utilizing Implicit and Explicit Relationships for Facial Attribute Classification (AAAI 2017 · similarity 0.88)
- Attributes Shape the Embedding Space of Face Recognition Models (ICML 2025 · similarity 0.88)
- MOON: A Mixed Objective Optimization Network for the Recognition of Facial Attributes (ECCV 2016 · similarity 0.87)
- Heterogeneous Face Attribute Estimation: A Deep Multi-Task Learning Approach (TPAMI 2017 · similarity 0.87)
- Robust RGB-D Face Recognition Using Attribute-Aware Loss (TPAMI 2019 · similarity 0.86)
- Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors (ICLR 2025 · similarity 0.84)
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.
