Summary of Learn2talk: 3d Talking Face Learns From 2d Talking Face, by Yixiang Zhuang et al.
Learn2Talk: 3D Talking Face Learns from 2D Talking Face
by Yixiang Zhuang, Baoping Cheng, Yao Cheng, Yuntao Jin, Renshuai Liu, Chengyang Li, Xuan Cheng, Jing Liao, Juncong Lin
First submitted to arxiv on: 19 Apr 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Graphics (cs.GR); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Learn2Talk framework aims to improve 3D talking face generation by leveraging expertise from 2D talking faces. It consists of a 3D sync-lip expert model for lip-syncing audio with facial motion and an audio-to-3D motions regression network guided by a teacher model selected from 2D talking face methods. This framework is evaluated extensively, showing advantages in terms of lip-sync, vertex accuracy, and speech perception compared to state-of-the-art models. The proposed framework has applications in audio-visual speech recognition and speech-driven 3D Gaussian Splatting based avatar animation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a new way to make 3D faces talk like real people. It uses ideas from making 2D faces talk and combines them with special techniques for lip-syncing and facial movement. The result is a more realistic and accurate 3D talking face that can be used in things like speech recognition and animations. |
Keywords
» Artificial intelligence » Regression » Teacher model
