3DiFACE: Synthesizing and Editing Holistic 3D Facial Animation (3DV 2025)
Автор: Justus Thies
Загружено: 2025-02-12
Просмотров: 555
We present 3DiFACE, a novel method for holistic speech-driven 3D facial animation. 3DiFACE produces diverse plausible lip and head motions for a single audio input and allows for editing via keyframing and interpolation. Specifically, we propose a fully-convolutional diffusion model that leverages the viseme-level diversity in our training corpus. Additionally, we employ a speaking-style personalization and a novel sparsely-guided motion diffusion to enable precise control and editing. Through quantitative and qualitative evaluations, we demonstrate that our method is capable of generating and editing diverse holistic 3D facial animations given a single audio input, with control between high fidelity and diversity.
Project page: https://balamuruganthambiraja.github....
Chapters:
0:00 Introduction
0:25 Editing demo, part-regeneration and keyframing
1:15 Existing works
1:52 Key-challenges
2:27 Our contribution
3:06 Editing using out-of-distribution motion and TTS
3:48 Holistic 3D animation comparison vs Baselines
4:29 Editing comparison vs Baselines
5:36 Style-personalisation comparison vs Baselines
5:56 Facial motion synthesis comparison
6:13 Diversity demonstration
6:28 Ablation
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке:
![NeuralFur: Animal Fur Reconstruction from Multi-view Images [3DV2026]](https://imager.clipsaver.ru/XshGOIKe_yU/max.jpg)
![MICA: Towards Metrical Reconstruction of Human Faces [ECCV2022]](https://imager.clipsaver.ru/vzzEbvv08VA/max.jpg)
![Im2Haicut: Single-view Strand-based Hair Reconstruction for Human Avatars [ICCV2026]](https://imager.clipsaver.ru/7CPsjuBW_10/max.jpg)