Joint Fine-Tuning in Deep Neural Networks for Facial Expression Recognition

By Heechul Jung, Sihaeng Lee, Junho Yim, Sunjeong Park, Junmo Kim. IEEE International Conference on Computer Vision, 2015

676 citations recorded by the source corpus.

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

View Paper: https://www.semanticscholar.org/paper/2911e7f0fb6803851b0eddf8067a6fc06e8eadd6

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

Modeling Stylized Character Expressions via Deep Learning (2016, methodology): “Joint Fine-Tuning in Deep Neural Networks [33] have also been used to combine temporal appearance features from image sequences and temporal geometry features from temporal facial landmark points to enhance the performance of the facial expression recognition.”
Region Attention Networks for Pose and Occlusion Robust Facial Expression Recognition (2019, background): “Due to its potential applications in various fields, such as intelligent tutoring systems, service robots, driver fatigue monitoring, Facial Expression Recognition (FER) has attracted increasing attention in the computer vision community recently [28], [5],”
Facial emotion recognition using convolutional neural networks (FERC) (2020, background): “Jung et al. [38] tried to work with fixed frames which make the system not so efficient with video input.”
Aff-Wild: Valence and Arousal ‘In-the-Wild’ Challenge (2017, methodology): “Some representative datasets, which are still used in many recent works [18], include the Cohn-Kanade database [33, 24], the MMI database [29, 35], the Multi-PIE database [17] and the BU-3D and BU-4D databases [41, 40].”
Privileged Information for Modeling Affect In The Wild (2021, methodology): “In [22], the authors use CNNs to learn geometric and temporal features describing facial action points to boost facial-based affect models’ performance.”
Debiasing Pre-Trained Language Models via Efficient Fine-Tuning (2022, neutral): “Fine-tuning, one approach of transfer learning, has been widely used for neural network models (Ge and Yu, 2017; Jung et al., 2015; Maqsood et al., 2019; Shin et al., 2016).”
A Brief Review of Facial Emotion Recognition Based on Visual Information (2018, methodology): “• Joint fine-tuning in deep neural networks [48] Sequence 70.”
IExpressNet: Facial Expression Recognition with Incremental Classes (2020, methodology): “To bridge the affective gap, various hand-crafted [6, 16, 63, 77] or learning-based features [24, 26, 71, 74] are designed.”
Facial Expression Recognition with Inconsistently Annotated Datasets (2018, background): “Based on these datasets, lots of FER approaches are proposed and achieve the state-of-the-art performance[27, 25, 30, 4, 14, 34, 18].”
A Novel Space-Time Representation on the Positive Semidefinite Cone for Facial Expression Recognition (2017, methodology): “Here again, the protocol used in [14, 20] was followed according to a 10-fold crossvalidation schema.”
Facial Expression Recognition Based on Deep Evolutional Spatial-Temporal Networks (2017, methodology): “of previous state-of-the-art methods [7], [6], [24] on this database is lower than the results on CK+.”
Deep Facial Expression Recognition: A Survey (2018, methodology): “In static-based methods [12], [13], [14], the feature representation is encoded with only spatial information from the current single image, whereas dynamicbased methods [15], [16], [17] consider the temporal relation among contiguous frames in the input facial expression sequence.”
Deep Affect Prediction in-the-Wild: Aff-Wild Database and Challenge, Deep Architectures, and Beyond (2018, methodology): “still used in many recent works [27], are the Cohn-Kanade database [34,54], MMI database [40,56], Multi-PIE database [22] and the BU-3D and BU-4D databases [61, 62].”
Reliable Crowdsourcing and Deep Locality-Preserving Learning for Unconstrained Facial Expression Recognition (2019, methodology): “In DTAGN [30], two different models were combined to extract temporal appearance and geometry features simultaneously.”
From Facial Expression Recognition to Interpersonal Relation Prediction (2016, background): “Social signal processing mainly involves facial expression recognition (Zhao and Pietikainen 2007; Tian et al. 2011; Liu et al. 2014b; Ruiz et al. 2015; Wu and Ji 2016; Fabian Benitez-Quiroz et al. 2016; Liu et al. 2014a, c; Jung et al. 2015;Mollahosseini et al. 2016; Fabian Benitez-Quiroz et al. 20”
Facial Expression Recognition by De-expression Residue Learning (2018, methodology): “Our DeRL method achieves the highest accuracy compared to those of state-of-the-art methods, including CNN-based methods (DTAGN-Joint [12], FN2EN [6], and PPDN [35]) and handcrafted features based methods (LBP-TOP [34], HOG 3D [14] and STM-Explet [18]).”