Lip Movements Generation at a Glance

By Lele Chen, Zhiheng Li, Ross K. Maddox, Zhiyao Duan, Chenliang Xu. European Conference on Computer Vision, 2018

324 citations recorded by the source corpus.

Cross-modality generation is an emerging topic that aims to synthesize data in one modality based on information in a different modality. In this paper, we consider a task of such: given an arbitrary audio speech and one lip image of arbitrary target identity, generate synthesized lip movements of the target identity saying the speech. To perform well in this task, it inevitably requires a model to not only consider the retention of target identity, photo-realistic of synthesized images, consistency and smoothness of lip images in a sequence, but more importantly, learn the correlations between audio speech and lip movements. To solve the collective problems, we explore the best modeling of the audio-visual correlations in building and training a lip-movement generator network. Specifically, we devise a method to fuse audio and image embeddings to generate multiple lip images at once and propose a novel correlation loss to synchronize lip changes and speech changes. Our final model utilizes a combination of four losses for a comprehensive consideration in generating lip movements; it is trained in an end-to-end fashion and is robust to lip shapes, view angles and different facial characteristics. Thoughtful experiments on three datasets ranging from lab-recorded to lips in-the-wild show that our model significantly outperforms other state-of-the-art methods extended to this task.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/d7e12f93339dc9a97cc325a4a3e9a13bdffb4988

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Deep Audio-visual Learning: A Survey (2020, background): “…For a lab-controlled environment only Talking Face Vougioukas et al. [98] Applied a temporal GAN for more temporal consistency Chen et al. [99] Applied optical ﬂow Generated lips only Zhou et al. [100] Disentangled information Lacked realism Zhu et al. [73] Asymmetric mutual information…”
What comprises a good talking-head video generation?: A Survey and Benchmark (2020, methodology): “Then we training the network with the metric learning step using ArcLoss, which leads LRS3-TED VoxCeleb2 Real Chen [5] Baseline Wang [17] Wiles [18] Zakharov [20] Real Chen [5] Baseline Wang [17] Wiles [18] Zakharov [20] Top-1 72.62% 1.99% 3.85% 2.23% 1.93% 1.77% 42.46% 2.40% 3.04% 1.64% 1.87% 1.99%”
FaceFormer: Speech-Driven 3D Facial Animation with Transformers (2021, methodology): “While aware of extensive 2D-based approaches [10, 11, 16, 18, 23, 29, 49, 59, 66, 70], we focus on animating a 3D model in this work.”
Capture, Learning, and Synthesis of 3D Speaking Styles (2019, background): “[16] correlate audio and image motion to synthesize lip motion of arbitrary identities.”
A Lip Sync Expert Is All You Need for Speech to Lip Generation In the Wild (2020, background): “Although LMD [4] focuses on the lip region, we found that lip landmarks can be quite inaccurate on generated faces.”
MakeItTalk: Speaker-Aware Talking Head Animation (2020, background): “[Chen et al. 2018] synthesized cropped lip movements for each frames.”
Pros and Cons of GAN Evaluation Measures: New Developments (2021, background): “Some other measures for video synthesis evaluation include Average Content Distance (Tulyakov et al., 2018) (the average L2 distance among all consecutive frames in a video), Cumulative Probability Blur Detection (CPBD) (Narvekar and Karam, 2009), Frequency Domain Blurriness Measure (FDBM) (De and M”
Realistic Speech-Driven Facial Animation with GANs (2019, methodology): “Subject independent approaches have been proposed that transform audio features to video frames [10, 8].”
Flow-guided One-shot Talking Face Generation with a High-resolution Audio-visual Dataset (2021, methodology): “While many works[7, 6, 18, 13, 51, 5, 48, 29, 42, 49, 4] make great efforts to synthesize realistic-looking videos, the generation of high-resolution videos is still a challenge.”
Hierarchical Cross-Modal Talking Face Generation With Dynamic Pixel-Wise Loss (2019, methodology): “Nonetheless, all the recent photo-realistic talking face generation methods [3, 12, 27, 1, 28, 35] failed to consider this problem.”
Pose-Controllable Talking Face Generation by Implicitly Modularized Audio-Visual Representation (2021, background): “…two different paths towards analyzing audio-driven talking human faces: 1) through pure latent feature learning and image reconstruction [14, 77, 9, 72, 50, 55, 44], and 2) to borrow the help of structural intermediate representations such as 2D landmarks [51, 10, 18] or 3D representations [1,…”
Learning Audio-Visual Speech Representation by Masked Multimodal Cluster Prediction (2022, unknown): “Additionally, it can benefit a vast range of applications, including but not limited to keyword spotting in sign language (Albanie et al., 2020), speech enhancement (Xu et al., 2020) and talking face generation (Chen et al., 2018).”