Cross-dataset learning and person-specific normalisation for automatic Action Unit detection

By Tadas Baltrusaitis, M. Mahmoud, Peter Robinson. IEEE International Conference on Automatic Face & Gesture Recognition, 2015

428 citations recorded by the source corpus.

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

View Paper: https://www.semanticscholar.org/paper/f472cb8380a41c540cfea32ebb4575da241c0288

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 Experimental and Cognitive Psychology

Citations in the corpus

AD-NeRF: Audio Driven Neural Radiance Fields for Talking Head Synthesis (2021, neutral): “…we propose a combined evaluation de-sign, including SyncNet [9] scores for audio-visual syn-7 chronization quality, Action Units (AU) detection [3] (by OpenFace [4]) for facial action coding consistency between source and generated results, and a diversiﬁed user study on image realism, ﬁdelity…”
Automatic Analysis of Facial Actions: A Survey (2019, methodology): “Some very recent works have given a spin to this idea and introduce a module predicting the neutral face at test time [13], [72].”
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.”
Psychology Meets Machine Learning: Interdisciplinary Perspectives on Algorithmic Job Candidate Screening (2018, methodology): “More specifically, regarding visual content, the open-source OpenFace library (Baltrušaitis et al. 2015) was used to detect and segment the face from frames in each video.”
Every Smile is Unique: Landmark-Guided Diverse Smile Generation (2018, methodology): “Indeed, we extract the intensity variations of these two AUs with the method in [3].”
The EMPATHIC Framework for Task Learning from Implicit Human Feedback (2020, methodology): “Leveraging existing models such as OpenFace [38, 39, 40] for extracting features alleviates our modeling burden with limited amount of data.”
Human-Explainable Features for Job Candidate Screening Prediction (2017, background): “OpenFace is an open source toolkit which does not only segment faces, but offers a feature extraction library that can extract and characterize facial movements and gaze [3].”
GANimation: Anatomically-aware Facial Animation from a Single Image (2018, methodology): “1 The dataset was re-annotated with [2] to obtain continuous activation annotations.”