Research map / Vision-language and multimodal / Cross-modal learning
A Quality-Guided Mixture of Score-Fusion Experts Framework for Human Recognition
Jie Zhu, Yiyang Su, Minchul Kim, Anil Jain, Xiaoming Liu
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 score triplet loss (vs triplet loss), one expert (ablation) | QME · MEVID | TAR@1%FAR: 23.3 -> 25.3 | Improved |
| add score triplet loss (vs triplet loss), two experts (ablation) | QME · MEVID | TAR@1%FAR: 30.8 -> 31.3 | Improved |
| add score triplet loss, one expert (ablation) | QME · MEVID | FNIR@1%FPIR: 84.0 -> 70.4 | Improved |
| increase experts from 1 to 2 (ablation) | QME · MEVID | TAR@1%FAR: 23.3 -> 30.8 | Improved |
| add QE-guided weighting of experts (ablation) | QME · MEVID | TAR@1%FAR: 31.3 -> 32.9 | Improved |
| add QE-guided weighting of experts (ablation) | QME · MEVID | Rank-1: 55.1 -> 55.7 | Improved |
| add QE-guided weighting of experts (ablation) | QME · MEVID | FNIR@1%FPIR: 66.5 -> 64.6 | Improved |
| combine both experts instead of expert 1 alone (ablation) | QME · BRIAR Face-Restricted | TAR@0.1%FAR: 62.0 -> 67.9 | Improved |
| combine both experts instead of expert 2 alone (ablation) | QME · BRIAR Face-Included | TAR@0.1%FAR: 81.8 -> 84.5 | Improved |
| use CAL for the quality estimator instead of AdaFace (ablation) | QME · CCVID | Rank-1: 92.6 -> 94.1 | Mixed |
| use CAL-QE instead of AdaFace-QE (ablation) | QME · MEVID | TAR@1%FAR: 32.9 -> 32.5 | No clear effect |
| add face modality to two body models (3-model vs 2-model fusion) | QME · LTCC | Rank-1: 75.3 -> 73.8 | Hurt |
Closest papers by meaning
- Flexible Biometrics Recognition: Bridging the Multimodality Gap through Attention Alignment and Prompt Tuning (CVPR 2024 · similarity 0.85)
- QAGait: Revisit Gait Recognition from a Quality Perspective (AAAI 2024 · similarity 0.83)
- Exploring More from Multiple Gait Modalities for Human Identification (AAAI 2025 · similarity 0.81)
- FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification (ICML 2025 · similarity 0.81)
- Heterogeneous Test-Time Training for Multi-Modal Person Re-identification (AAAI 2024 · similarity 0.81)
- FuseMoE: Mixture-of-Experts Transformers for Fleximodal Fusion (NeurIPS 2024 · similarity 0.80)
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.
