MEAD: A Large-Scale Audio-Visual Dataset for Emotional Talking-Face Generation

By Kaisiyuan Wang, Qianyi Wu, Linsen Song, Zhuoqian Yang, Wayne Wu, Chen Qian, Ran He, Yu Qiao, Chen Change Loy. European Conference on Computer Vision, 2020

491 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/ce752f69e167db27d4c186f5dfcd064e9b6e7f3b

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

1. Chen Change Loy
2. Luc Van Gool
3. Ying Shan
4. Xintao Wang
5. Ziwei Liu
6. Dacheng Tao
7. Ming-Hsuan Yang
8. Yu Qiao

In Signal Processing

1. Chen Change Loy
2. Ran He
3. Xintao Wang
4. Wayne Wu
5. Ying Shan
6. Mandi Luo
7. Hao Zhu
8. Rui Wang

Citations in the corpus

• emotion2vec: Self-Supervised Pre-Training for Speech Emotion Representation (2023, methodology): “In the pretraining phase, we utilize five large-scale English datasets, including IEMO-CAP (Busso et al., 2008), MELD (Poria et al., 2019), MEAD (Wang et al., 2020), CMU-MOSEI (Zadeh et al., 2018), and MSP-Podcast (Martinez-Lucas et al., 2020), resulting in a total of 262 hours.”
• EMOCA: Emotion Driven Monocular Face Capture and Animation (2022, background): “2 [93] Kaisiyuan Wang, Qianyi Wu, Linsen Song, Zhuoqian Yang, Wayne Wu, Chen Qian, Ran He, Yu Qiao, and Chen Change Loy.”
• Deep Audio-visual Learning: A Survey (2020, background): “Classes Length* Year GRID [119] Lab 34 33,000 2006 Lombard Grid [120] Lab 54 54,000 2018 TCD TIMIT [121] Lab 62 - 2015 Vid TIMIT [122] Lab 43 - 2009 RAVDESS [123] Lab 24 - 2018 SEWA [124] Lab 180 - 2017 Speech OuluVS [125] Lab 20 1000 2009 OuluVS2 [126] Lab 52 3640 2016 Voxceleb [127] Wild 1,251 154”
• PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding (2023, background): “However, existing human datasets either do not annotate ID information [31, 37, 61, 78], or the richness of the scenes they contain is very limited [38, 46, 68] ( i.e ., they only focus on the face area).”
• AD-NeRF: Audio Driven Neural Radiance Fields for Talking Head Synthesis (2021, methodology): “Therefore, most of these meth-ods utilize some intermediate face representations including reconstructing explicit 3D face shapes [55] and regressing expression coefficients [43] or 2D landmarks [41, 47].”
• Flow-guided One-shot Talking Face Generation with a High-resolution Audio-visual Dataset (2021, background): “5 51 RAVDESS[24] Lab 2018 1280×1024 24 7 8 MEAD[43] Lab 2020 1920×1080 60 40 159 Our HDTF Wild 2020 720P∼1080P 300+ 15.”
• StyleGAN-V: A Continuous Video Generator with the Price, Image Quality and Perks of StyleGAN2 (2021, unknown): “Generations on MEAD 1024(2) [72] for MoCoGAN-HD [65] and our method.”
• CelebV-HQ: A Large-Scale Video Facial Attributes Dataset (2022, methodology): “We construct a benchmark of unconditional video generation task, for four currently prevalent models (VideoGPT [90], MoCoGAN-HD [75], DIGAN [94], and StyleGAN-V [73]) on 4 face video datasets (FaceForensics [65], Vox [59], MEAD [82] and CelebV-HQ).”