Recognizing Action Units for Facial Expression Analysis

By Ying-li Tian, Takeo Kanade, Jeffrey F. Cohn. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001

1,974 citations recorded by the source corpus.

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

View Paper: https://www.semanticscholar.org/paper/0c5afb209b647456e99ce42a6d9d177764f9a0dd

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

Eyeblink-based Anti-Spoofing in Face Recognition from a Generic Webcamera (2007, background): “Most of previous efforts need highly controlled conditions and high-quality input data, for example, the system of automatic recognition of human facial action units [16].”
Detecting depression from facial actions and vocal prosody (2009, background): “Tian, Kanade, & Cohn [ 1 ], for example, discriminated between 34 posed action units and action unit combinations.”
Knowledge Augmented Deep Neural Networks for Joint Facial Expression and Action Unit Recognition (2020, background): “In addition, Tian et al [24] proposed to employ AU co-occurrence results for facial expression recognition.”
Deep Learning for Emotion Recognition on Small Datasets using Transfer Learning (2015, methodology): “More detailed approaches follow the Facial Action Coding System (FACS), attempting either to classify which Action Units (AU) are activated [24] or to estimate their intensity [18].”
Discovering Motion Primitives for Unsupervised Grouping and One-Shot Learning of Human Actions, Gestures, and Expressions (2013, positive): “We tested our hypothesis on the Cohn-Kanade AU-Coded Facial Expression Database [40], which contains videos of multiple humans depicting various expressions corresponding to human emotions.”
IExpressNet: Facial Expression Recognition with Incremental Classes (2020, background): “To bridge the affective gap, various hand-crafted [6, 16, 63, 77] or learning-based features [24, 26, 71, 74] are designed.”
The first facial expression recognition and analysis challenge (2011, background): “…of existing automatic facial expression recognition literature can be divided based on the types of features they use in three ways: those that use appearance-based features (e.g. [5], [2]), those that use geometric feature-based approaches (e.g. [17], [7]), and those that use both (e.g. [15]).”
Gesture Recognition: A Survey (2007, methodology): “For example, in FACS, an " inner brow raiser " corresponds to AU1 while a " jaw drop " refers to AU26.”
A real-time automated system for the recognition of human facial expressions (2006, methodology): “A commonly used approach to improve robustness in classifying expression is to combine the results of several different methods [4], [20], [21].”
Active and dynamic information fusion for facial expression understanding from image sequences (2005, background): “The works in [36] and [22] focus on recognizing facial AUs rather than facial expressions.”
DISFA: A Spontaneous Facial Action Intensity Database (2013, background): “The AUs chosen were among those most common in emotion expression and social interaction and that have been studied previously in computer vision and machine learning [45].”
2D vs. 3D Deformable Face Models: Representational Power, Construction, and Real-Time Fitting (2007, neutral): “M̃2i+2 Qk M̃ 2 j+2 = { 1 if i = j = k 0 if i = k (24)”
Facial expression recognition from near-infrared videos (2011, background): “4 (b), since most of the facial features are concentrated on mouth, cheeks, eyes and forehead [32].”
Dynamics of Facial Expression Extracted Automatically from Video (2004, methodology): “Three groups besides ours have focused on automatic FACS recognition as a tool for behavioral research:[22, 19, 12].”
From Facial Expression Recognition to Interpersonal Relation Prediction (2016, methodology): “[71] use the positions of facial landmarks for facial action recognition and then perform expression analysis.”
Every Smile is Unique: Landmark-Guided Diverse Smile Generation (2018, methodology): “To do that, we rely on action Units (AUs) [37] and specifically on the cheek raiser and lip corner puller AUs.”
Face Expression Recognition and Analysis: The State of the Art (2012, methodology): “An example of such a system is the Tian et al.’s AFA system that we saw in section 3.1 [10].”
3D Facial Expression Recognition Based on Primitive Surface Feature Distribution (2006, background): “, frowns or wrinkles) [22] or action units [8, 6].”