Bilingual Text-to-Motion Generation: A New Benchmark and Baselines
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914424371019776 |
|---|---|
| author | Weng, Wanjiang Tan, Xiaofeng Shu, Xiangbo Xie, Guo-Sen Zhou, Pan Wang, Hongsong |
| author_facet | Weng, Wanjiang Tan, Xiaofeng Shu, Xiangbo Xie, Guo-Sen Zhou, Pan Wang, Hongsong |
| contents | Text-to-motion generation holds significant potential for cross-linguistic applications, yet it is hindered by the lack of bilingual datasets and the poor cross-lingual semantic understanding of existing language models. To address these gaps, we introduce BiHumanML3D, the first bilingual text-to-motion benchmark, constructed via LLM-assisted annotation and rigorous manual correction. Furthermore, we propose a simple yet effective baseline, Bilingual Motion Diffusion (BiMD), featuring Cross-Lingual Alignment (CLA). CLA explicitly aligns semantic representations across languages, creating a robust conditional space that enables high-quality motion generation from bilingual inputs, including zero-shot code-switching scenarios. Extensive experiments demonstrate that BiMD with CLA achieves an FID of 0.045 vs. 0.169 and R@3 of 82.8\% vs. 80.8\%, significantly outperforms monolingual diffusion models and translation baselines on BiHumanML3D, underscoring the critical necessity and reliability of our dataset and the effectiveness of our alignment strategy for cross-lingual motion synthesis. The dataset and code are released at \href{https://wengwanjiang.github.io/BilingualT2M-page}{https://wengwanjiang.github.io/BilingualT2M-page} |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_25178 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Bilingual Text-to-Motion Generation: A New Benchmark and Baselines Weng, Wanjiang Tan, Xiaofeng Shu, Xiangbo Xie, Guo-Sen Zhou, Pan Wang, Hongsong Computer Vision and Pattern Recognition Computation and Language Text-to-motion generation holds significant potential for cross-linguistic applications, yet it is hindered by the lack of bilingual datasets and the poor cross-lingual semantic understanding of existing language models. To address these gaps, we introduce BiHumanML3D, the first bilingual text-to-motion benchmark, constructed via LLM-assisted annotation and rigorous manual correction. Furthermore, we propose a simple yet effective baseline, Bilingual Motion Diffusion (BiMD), featuring Cross-Lingual Alignment (CLA). CLA explicitly aligns semantic representations across languages, creating a robust conditional space that enables high-quality motion generation from bilingual inputs, including zero-shot code-switching scenarios. Extensive experiments demonstrate that BiMD with CLA achieves an FID of 0.045 vs. 0.169 and R@3 of 82.8\% vs. 80.8\%, significantly outperforms monolingual diffusion models and translation baselines on BiHumanML3D, underscoring the critical necessity and reliability of our dataset and the effectiveness of our alignment strategy for cross-lingual motion synthesis. The dataset and code are released at \href{https://wengwanjiang.github.io/BilingualT2M-page}{https://wengwanjiang.github.io/BilingualT2M-page} |
| title | Bilingual Text-to-Motion Generation: A New Benchmark and Baselines |
| topic | Computer Vision and Pattern Recognition Computation and Language |
| url | https://arxiv.org/abs/2603.25178 |