publications

2025

1. FilmComposer: LLM-Driven Music Production for Silent Film Clips

   Zhifeng Xie ,  Qile He ,  Youjia Zhu ,  Qiwei He ,  and  Mengtian Li*

   In CVPR , 2025

   Bib Website

   @inproceedings{li2025filmcomposer,
     title = {FilmComposer: LLM-Driven Music Production for Silent Film Clips},
     author = {Xie, Zhifeng and He, Qile and Zhu, Youjia and He, Qiwei and Li*, Mengtian},
     booktitle = {CVPR},
     year = {2025},
     publisher = {IEEE},
   }

2. StageDesigner: Artistic Stage Generation for Scenography via Theater Scripts

   Zhaoxing Gan ,  Mengtian Li* ,  Ruhua Chen ,  Zhongxia Ji ,  Sichen Guo ,  Huanling Hu ,  Guangnan Ye ,  and  Zuo Hu

   In CVPR , 2025

   Bib Website

   @inproceedings{li2025stagedesigner,
     title = {StageDesigner: Artistic Stage Generation for Scenography via Theater Scripts},
     author = {Gan, Zhaoxing and Li*, Mengtian and Chen, Ruhua and Ji, Zhongxia and Guo, Sichen and Hu, Huanling and Ye, Guangnan and Hu, Zuo},
     booktitle = {CVPR},
     year = {2025},
     publisher = {IEEE},
   }

3. AniGaussian: Animatable Gaussian Avatar with Pose-guided Deformation

   Mengtian Li ,  Shengxiang Yao ,  Chen Kai ,  Zhifeng Xie ,  Keyu Chen ,  and  Yu-Gang Jiang

   2025

   Bib

   @misc{li2025anigaussian,
     title = {AniGaussian: Animatable Gaussian Avatar with Pose-guided Deformation},
     author = {Li, Mengtian and Yao, Shengxiang and Kai, Chen and Xie, Zhifeng and Chen, Keyu and Jiang, Yu-Gang},
     journal = {arXiv preprint arXiv:2502.19441},
     year = {2025},
     primaryclass = {cs.CV}
   }

4. LMTalker: Sparse Landmark-guided Gaussian Splatting for High-fidelity Talking Head Synthesis

   Zhifeng Xie ,  Zhiwen Jiang ,  Xuemin Lei ,  and  Mengtian Li*

   In ICASSP , 2025

   Bib Website

   @inproceedings{li2024lmtalker,
     title = {LMTalker: Sparse Landmark-guided Gaussian Splatting for High-fidelity Talking Head Synthesis},
     author = {Xie, Zhifeng and Jiang, Zhiwen and Lei, Xuemin and Li*, Mengtian},
     booktitle = {ICASSP},
     year = {2025},
     pages = {1--5},
     publisher = {IEEE},
   }

5. Knowledge Transfer Across Modalities for Weakly Supervised Point Cloud Semantic Segmentation

   Zihan Wang ,  Yunhang Shen ,  Mengtian Li ,  Ke Li ,  Xing Sun ,  Shaohui Lin ,  and  Lizhuang Ma

   In ICASSP , 2025

   Bib

   @inproceedings{wang2025knowledge,
     title = {Knowledge Transfer Across Modalities for Weakly Supervised Point Cloud Semantic Segmentation},
     author = {Wang, Zihan and Shen, Yunhang and Li, Mengtian and Li, Ke and Sun, Xing and Lin, Shaohui and Ma, Lizhuang},
     booktitle = {ICASSP},
     pages = {1--5},
     year = {2025},
     organization = {IEEE}
   }

2024

1. ArtNVG: Content-Style Separated Artistic Neighboring-View Gaussian Stylization

   Zixiao Gu ,  Mengtian Li* ,  Ruhua Chen ,  Zhongxia Ji ,  Sichen Guo ,  Zhenye Zhang ,  Guangnan Ye ,  and  Zuo Hu

   2024

   Bib

   @misc{li2024artnvg,
     title = {ArtNVG: Content-Style Separated Artistic Neighboring-View Gaussian Stylization},
     author = {Gu, Zixiao and Li*, Mengtian and Chen, Ruhua and Ji, Zhongxia and Guo, Sichen and Zhang, Zhenye and Ye, Guangnan and Hu, Zuo},
     journal = {arXiv preprint arXiv:2412.18783},
     year = {2024},
     primaryclass = {cs.CV}
   }

2. HieraFashDiff: Hierarchical Fashion Design with Multi-stage Diffusion Models

   Zhifeng Xie ,  Hao Li ,  Huiming Ding ,  Mengtian Li ,  Xinhan Di ,  and  Ying Cao

   In AAAI , 2024

   Bib Website

   @inproceedings{li2024hierahashdiff,
     title = {HieraFashDiff: Hierarchical Fashion Design with Multi-stage Diffusion Models},
     author = {Xie, Zhifeng and Li, Hao and Ding, Huiming and Li, Mengtian and Di, Xinhan and Cao, Ying},
     booktitle = {AAAI},
     year = {2024},
     publisher = {AAAI},
   }

3. Infinite Motion: Extended Motion Generation via Long Text Instructions

   Mengtian Li ,  Chengshuo Zhai ,  Shengxiang Yao ,  Zhifeng Xie ,  Keyu Chen ,  and  Yu-Gang Jiang

   2024

   Bib Website

   @misc{li2024infinite,
     title = {Infinite Motion: Extended Motion Generation via Long Text Instructions},
     author = {Li, Mengtian and Zhai, Chengshuo and Yao, Shengxiang and Xie, Zhifeng and Chen, Keyu and Jiang, Yu-Gang},
     eprint = {2407.08443},
     archiveprefix = {arXiv},
     year = {2024},
     primaryclass = {cs.CV},
   }

4. SonicVisionLM: Playing Sound with Vision Language Models

   Zhifeng Xie ,  Shengye Yu ,  Qile He ,  and  Mengtian Li*

   In CVPR , 2024

   Bib Website

   @inproceedings{xie2024sonicvisionlm,
     title = {SonicVisionLM: Playing Sound with Vision Language Models},
     author = {Xie, Zhifeng and Yu, Shengye and He, Qile and Li*, Mengtian},
     booktitle = {CVPR},
     pages = {14930--14939},
     year = {2024},
   }

5. GaussianBody: Clothed Human Reconstruction via 3d Gaussian Splatting

   Mengtian Li ,  Shengxiang Yao ,  Zhifeng Xie ,  and  Keyu Chen

   2024

   Bib

   @misc{li2024gaussianbody,
     title = {GaussianBody: Clothed Human Reconstruction via 3d Gaussian Splatting},
     author = {Li, Mengtian and Yao, Shengxiang and Xie, Zhifeng and Chen, Keyu},
     year = {2024},
     eprint = {2401.09720},
     archiveprefix = {arXiv},
     primaryclass = {cs.CV}
   }

6. Class-imbalanced semi-supervised learning for large-scale point cloud semantic segmentation via decoupling optimization

   Mengtian Li ,  Shaohui Lin ,  Zihan Wang ,  Yunhang Shen ,  Baochang Zhang ,  and  Lizhuang Ma

   Pattern Recognition, 2024

   Bib

   @article{LI2024110701,
     title = {Class-imbalanced semi-supervised learning for large-scale point cloud semantic segmentation via decoupling optimization},
     journal = {Pattern Recognition},
     volume = {156},
     pages = {110701},
     year = {2024},
     issn = {0031-3203},
     doi = {https://doi.org/10.1016/j.patcog.2024.110701},
     url = {https://www.sciencedirect.com/science/article/pii/S0031320324004527},
     author = {Li, Mengtian and Lin, Shaohui and Wang, Zihan and Shen, Yunhang and Zhang, Baochang and Ma, Lizhuang}
   }

2023

1. A fine-grained vision and language representation framework with graph-based fashion semantic knowledge

   Huiming Ding ,  Sen Wang ,  Zhifeng Xie ,  Mengtian Li* ,  and  Lizhuang Ma

   Computers & Graphics, 2023

   Bib

   @article{ding2023fine,
     title = {A fine-grained vision and language representation framework with graph-based fashion semantic knowledge},
     author = {Ding, Huiming and Wang, Sen and Xie, Zhifeng and Li*, Mengtian and Ma, Lizhuang},
     journal = {Computers \& Graphics},
     volume = {115},
     pages = {216--225},
     year = {2023},
     publisher = {Elsevier}
   }

2022

1. Hyperspherical learning in multi-label classification

   Bo Ke ,  Yunquan Zhu ,  Mengtian Li ,  Xiujun Shu ,  Ruizhi Qiao ,  and  Bo Ren

   In ECCV , 2022

   Bib

   @inproceedings{ke2022hyperspherical,
     title = {Hyperspherical learning in multi-label classification},
     author = {Ke, Bo and Zhu, Yunquan and Li, Mengtian and Shu, Xiujun and Qiao, Ruizhi and Ren, Bo},
     booktitle = {ECCV},
     pages = {38--55},
     year = {2022},
     organization = {Springer}
   }

2. Hybridcr: Weakly-supervised 3d point cloud semantic segmentation via hybrid contrastive regularization

   Mengtian Li ,  Yuan Xie ,  Yunhang Shen ,  Bo Ke ,  Ruizhi Qiao ,  Bo Ren ,  Shaohui Lin ,  and  Lizhuang Ma

   In CVPR , 2022

   Bib

   @inproceedings{li2022hybridcr,
     title = {Hybridcr: Weakly-supervised 3d point cloud semantic segmentation via hybrid contrastive regularization},
     author = {Li, Mengtian and Xie, Yuan and Shen, Yunhang and Ke, Bo and Qiao, Ruizhi and Ren, Bo and Lin, Shaohui and Ma, Lizhuang},
     booktitle = {CVPR},
     pages = {14930--14939},
     year = {2022}
   }

3. Paying attention for adjacent areas: Learning discriminative features for large-scale 3D scene segmentation

   Mengtian Li ,  Yuan Xie ,  and  Lizhuang Ma

   Pattern Recognition, 2022

   Bib

   @article{li2022paying,
     title = {Paying attention for adjacent areas: Learning discriminative features for large-scale 3D scene segmentation},
     author = {Li, Mengtian and Xie, Yuan and Ma, Lizhuang},
     journal = {Pattern Recognition},
     volume = {129},
     pages = {108722},
     year = {2022},
     publisher = {Elsevier}
   }