XM-ALIGN: Unified Cross-Modal Embedding Alignment for Face-Voice Association
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913068589514752 |
|---|---|
| author | Fang, Zhihua Tao, Shumei Wang, Junxu He, Liang |
| author_facet | Fang, Zhihua Tao, Shumei Wang, Junxu He, Liang |
| contents | This paper introduces our solution, XM-ALIGN (Unified Cross-Modal Embedding Alignment Framework), proposed for the FAME challenge at ICASSP 2026. Our framework combines explicit and implicit alignment mechanisms, significantly improving cross-modal verification performance in both "heard" and "unheard" languages. By extracting feature embeddings from both face and voice encoders and jointly optimizing them using a shared classifier, we employ mean squared error (MSE) as the embedding alignment loss to ensure tight alignment between modalities. Additionally, data augmentation strategies are applied during model training to enhance generalization. Experimental results show that our approach demonstrates superior performance on the MAV-Celeb dataset. The code will be released at https://github.com/PunkMale/XM-ALIGN. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_06757 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | XM-ALIGN: Unified Cross-Modal Embedding Alignment for Face-Voice Association Fang, Zhihua Tao, Shumei Wang, Junxu He, Liang Sound Computer Vision and Pattern Recognition This paper introduces our solution, XM-ALIGN (Unified Cross-Modal Embedding Alignment Framework), proposed for the FAME challenge at ICASSP 2026. Our framework combines explicit and implicit alignment mechanisms, significantly improving cross-modal verification performance in both "heard" and "unheard" languages. By extracting feature embeddings from both face and voice encoders and jointly optimizing them using a shared classifier, we employ mean squared error (MSE) as the embedding alignment loss to ensure tight alignment between modalities. Additionally, data augmentation strategies are applied during model training to enhance generalization. Experimental results show that our approach demonstrates superior performance on the MAV-Celeb dataset. The code will be released at https://github.com/PunkMale/XM-ALIGN. |
| title | XM-ALIGN: Unified Cross-Modal Embedding Alignment for Face-Voice Association |
| topic | Sound Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.06757 |