# All's well that FID's well? Result quality and metric scores in GAN models for lip-sychronization tasks

arXiv: https://arxiv.org/abs/2212.13810 · HTML: https://mlsanity.com/paper/2212.13810/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 23 references that are in the map.
- First posted on arXiv 28 Dec 2022.

## Builds on (most cited first, 13 of 13)

- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Face2Face: Real-time Face Capture and Reenactment of RGB Videos](https://mlsanity.com/paper/2007.14808/) (2020)
- [Deep Audio-Visual Speech Recognition](https://mlsanity.com/paper/1809.02108/) (2018)
- [Talking Face Generation by Adversarially Disentangled Audio-Visual Representation](https://mlsanity.com/paper/1807.07860/) (2018)
- [Video Frame Synthesis using Deep Voxel Flow](https://mlsanity.com/paper/1702.02463/) (2017)
- [X2Face: A network for controlling face generation by using images, audio, and pose codes](https://mlsanity.com/paper/1807.10550/) (2018)
- [Effectively Unbiased FID and Inception Score and where to find them](https://mlsanity.com/paper/1911.07023/) (2019)
- [Lip Movements Generation at a Glance](https://mlsanity.com/paper/1803.10404/) (2018)
- [Realistic Speech-Driven Facial Animation with GANs](https://mlsanity.com/paper/1906.06337/) (2019)
- [Towards Automatic Face-to-Face Translation](https://mlsanity.com/paper/2003.00418/) (2020)
- [An empirical study on evaluation metrics of generative adversarial networks](https://mlsanity.com/paper/1806.07755/) (2018)
- [ObamaNet: Photo-realistic lip-sync from text](https://mlsanity.com/paper/1801.01442/) (2017)
- [Iterative Text-based Editing of Talking-heads Using Neural Retargeting](https://mlsanity.com/paper/2011.10688/) (2020)

## Abstract (arXiv)

We test the performance of GAN models for lip-synchronization. For this, we reimplement LipGAN in Pytorch, train it on the dataset GRID and compare it to our own variation, L1WGAN-GP, adapted to the LipGAN architecture and also trained on GRID.