OpenFace 2.0: Facial Behavior Analysis Toolkit

By Tadas Baltrusaitis, Amir Zadeh, Yao Chong Lim, Louis-philippe Morency. 2018

1,882 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Experimental and Cognitive Psychology

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. Siwei Lyu
2. Hany Farid
3. Maja Pantic
4. Luisa Verdoliva
5. Yuezun Li
6. Michel François Valstar
7. Shruti Agarwal
8. Luc Van Gool

In Experimental and Cognitive Psychology

1. Maja Pantic
2. Björn Schuller
3. Michel François Valstar
4. R. Cowie
5. Mohammad Soleymani
6. Fabien Ringeval
7. Shruti Agarwal
8. Hao Li

Citations in the corpus

• FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age (2019, background): “Using off-the-shelf tools [2, 4] and commercial services, social scientists, who traditionally didn’t use images, begun to use images of people to infer their demographic attributes and analyze their behaviors in many studies.”
• AD-NeRF: Audio Driven Neural Radiance Fields for Talking Head Synthesis (2021, methodology): “SyncNet score [9] AU error [4 1.”
• AVEC 2019 Workshop and Challenge: State-of-Mind, Detecting Depression with AI, and Cross-Cultural Affect Recognition (2019, methodology): “Face toolkit [4] was exploited, which provided the best results on the test set for both arousal and valence with FAUs based features.”
• Exposing Deep Fakes Using Inconsistent Head Poses (2018, methodology): “(2) Then, with the standard 3D facial landmark model of the same 68 points from OpenFace2 [17], the head poses from central face region (Rc and tc) and whole face (Ra and ta) are estimated with landmarks 18 − 36, 49, 55 (red in Fig.”
• Interpretable and Trustworthy Deepfake Detection via Dynamic Prototypes (2021, background): “Early face forensic work focus on hand-crafting facial features, such as eye color and missing reflections [41], 3D head poses [70], and facial movements [2, 6].”
• DeepFakes and Beyond: A Survey of Face Manipulation and Fake Detection (2020, methodology): “For the feature extraction, the OpenFace2 toolkit was considered [76], obtaining an intensity and occurrence for 18 different facial action units related to movements of facial muscles such as cheek raiser, nose wrinkler, mouth stretch, etc. Additionally, four features related to head movements…”
• Protecting World Leaders Against Deep Fakes (2019, methodology): “We use the open-source facial behavior analysis toolkit OpenFace2 [3, 2, 4] to extract facial and head movements in a video.”
• FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age for Bias Measurement and Mitigation (2021, background): “Using off-the-shelf tools [2, 4] and commercial services, social scientists have begun to use images of people to infer their demographic attributes and analyze their behaviors.”