2026年 04期

摘要(Abstract)：

为了解决现有面部迁移方法在三维虚拟角色人脸模型中应用受限以及未充分考虑场景和姿态影响的问题，提出一种基于情绪感知和面部动作编码系统的虚拟角色面部迁移方法。首先设计基于卷积神经网络的面部特征迁移模块，从输入的人脸图像中提取关键特征点信息并计算参数，迁移至虚拟角色的面部模型；其次，提出一种融合姿态信息和场景语义的情绪感知方法，利用多层感知机模型对融合人体姿态和场景语义信息的虚拟角色情绪进行识别；最后，提出一种基于面部动作编码系统的情绪迁移方法，将识别出的情绪映射至相应的动作单元组合，实现虚拟角色的面部迁移。数据和仿真实验结果表明，所提方法能够实现表情的逼真、自然和多样化表现，并在复杂场景下生成与场景和姿态相匹配的自然表情，显著提升虚拟角色的情绪表达能力和用户的沉浸感。

关键词(KeyWords)：虚拟角色; 面部迁移; 情绪感知; 面部动作编码系统

基金项目(Foundation):国家自然科学基金项目(62172265, 62503299); 山东省教育科学规划重点课题(2020KZD003); 山东省自然科学基金项目(ZR2025MS1025, ZR2025QC665)

作者(Author):刘一良，李念梓，洪潇，陆佃杰，张桂娟

DOI:10.13349/j.cnki.jdxbn.20260514.002

参考文献(References)：

[1] Mystakidis S. Metaverse [J]. Encyclopedia, 2022, 2(1): 486.

[2] Lyytinen K, Nickerson J V, King J L. Metahuman systems = humans+machines that learn [J]. Journal of Information Technology, 2021, 36(4): 427.

[3] 黄进展, 文庭孝. 数字人体信息聚合: 内涵阐释、数据来源、模型构建与应用前景 [J]. 图书馆论坛, 2024, 44(7): 28.

[4] 娄方园, 齐梦娜, 王竹新, 等. 元宇宙场域下的教育数字人及其应用 [J]. 图书馆论坛, 2023, 43(3): 101.

[5] Lee J, Joo H. Mocap everyone everywhere: lightweight motion capture with smartwatches and a head-mounted camera[C] //Proceedings of the IEEE / CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 16-22, 2024, Seattle, WA, USA. Piscataway: IEEE Computer Society, 2024: 1091.

[6] Adlou B, Wilburn C, Weimar W. Motion capture technologies for athletic performance enhancement and injury risk assessment: a review for multi-sport organizations [J]. Sensors, 2025, 25( 14): 4384.

[7] Zhu Yuhao, Li Qi, Wang Jian, et al. One-shot face swapping on megapixels[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 20-25, 2021, Nashville, TN, USA. Piscataway: IEEE Computer Society, 2021: 4834.

[8] Tripathy S, Kannala J, Rahtu E. Icface: interpretable and controllable face reenactment using gans[C]//Proceedings of the IEEE / CVF Winter Conference on Applications of Computer Vision (WACV), March 1 - 5, 2020, Snowmass Village, CO, USA. Piscataway: IEEE Computer Society, 2020: 3385.

[9] Zakharov E, Ivakhnenko A, Shysheya A, et al. Fast bi-layer neural synthesis of one-shot realistic head avatars[C]//European Conference on Computer Vision, August 23-28, 2020, Online. Ber lin: Springer, 2020: 524.

[10] Lin Jiangke, Yuan Yi, Shao Tianjia, et al. Towards high-fidelity 3d face reconstruction from in-the-wild images using graph convolutional networks[C]//Proceedings of the IEEE / CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 13- 19, 2020, Seattle, WA, USA. Piscataway: IEEE Computer Society, 2020: 5891.

[11] Daněcˇek R, Black M J, Bolkart T. Emoca: emotion-driven monocular face capture and animation[C]//Proceedings of the IEEE / CVF Conference on Computer Vision and Pattern v (CVPR), June 18- 24, 2022, New Orleans, LA, USA. Piscataway: IEEE Computer Society, 2022: 20311.

[12] Feng Jinsong, Singhal P. 3D face style transfer with a hybrid solution of NeRF and mesh rasterization[C]//Proceedings of the IEEE/ CVF Winter Conference on Applications of Computer Vision (WACV), January 3-8, 2024, Waikoloa, HI, USA. Piscataway: IEEE Computer Society, 2024: 3504.

[13] Kang S, Song H, Yoon B, et al. The Influence of emotion-based prioritized facial expressions on social presence in avatar-mediated remote communication[C]//Proceedings of the IEEE International Symposium on Mixed and Augmented Reality (ISMAR), October 21-25, 2024, Bellevue, WA, USA. Piscataway: IEEE Computer Society, 2024: 1147.

[14] Ekman P, Friesen W V. Facial action coding system: a technique for the measurement of facial movement[M]. Palo Alto: Consulting Psychologists Press, 1978.

[15] King D E. Dlib-ml: a machine learning toolkit[J]. The Journal of Machine Learning Research, 2009, 10: 1755.

[16] Soukup ova T, Cech J. Eye blink detection using facial landmarks[C]//Proceedings of the 21st Computer Vision Winter Workshop (CVWW), February 3 - 5, 2016, Rimske Toplice, Slovenia. Piscataway: IEEE Computer Society, 2016: 2.

[17] Unity Asset Store. Avatar Maker Pro-3D avatar from a single selfie [EB/OL]. [2026-05-07]. https://assetstore.unity.com/ pack ages/ tools/ modeling / avatar-maker-pro-3d-avatar-from-a-single-se lfie-134800 / .

[18] 冯桑, 杨润彬, 黄懿, 等. 基于改进残差网络的驾驶员情绪监测方法[J]. 中国科技论文, 2024, 19(8): 945.

[19] Generated Photos[ EB / OL]. [2026- 05- 07]. https://genera ted. photos.

[20] Yang Jingyuan, Huang Qirui, Ding Tingting, et al. Emoset: a large-scale visual emotion dataset with rich attributes[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision(ICCV), October 1-6, 2023, Paris, France. Piscataway: IEEE Computer Society, 2023: 20383.

[21] Kosti R, Alvarez J M, Recasens A, et al. Emotic: emotions in context dataset [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), July 21 - 26, 2017, Honolulu, HI, USA. Piscataway: IEEE Computer Society, 2017: 61.

[22] Render people[EB/OL]. [2026 - 05 - 07]. https://render people. com.

[23] Liang Feng, Wu Bichen, Dai Xiaoliang, et al. Open-vocabulary semantic segmentation with mask-adapted clip[C]//Proceedings of the IEEE/ CVF Conference on Computer Vision and Pattern Recognition ( CVPR), June 17 - 24, 2023, Vancouver, BC, Canada. Piscataway: IEEE Computer Society, 2023: 7061.

[24] Sketchfab[EB/OL]. [2026-05-07]. https://sketchfab. com.

[25] 黄东晋, 俞乐洋, 石永生, 等. 基于情绪一致性的三维人脸重建 [J]. 上海大学学报(自然科学版), 2025, 31(3): 475.