Active and dynamic information fusion for facial expression understanding from image sequences

By Yongmian Zhang, Qiang Ji. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005

506 citations recorded by the source corpus.

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

View Paper: https://www.semanticscholar.org/paper/b04a9b4a6c21d1d394d82b26728e5cba5e7cc3dc

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

3D Facial Expression Recognition Based on Primitive Surface Feature Distribution (2006, background): “Most research over the past thirty years has been directed towards static/dynamic, analytic, 2D feature extraction [9], focusing primarily on two types of features: 2D geometric features and appearance features from 2D static images [16, 18] or video sequences [4, 3, 31].”
Facial expression recognition from near-infrared videos (2011, methodology): “Zhang and Ji [40] used 26 facial points around the eyes, eyebrows, andmouth, and the transient features proposedbyTian et al.”
Facial Expression Analysis under Partial Occlusion (2018, methodology): “Few studies have tested occlusion in brows [Miyakoshi and Kato 2011] or arising from hair [Xia, et al. 2009] and blur [Tan Dat and Ranganath 2008],[Tan Dat and Surendra 2008], as well as missed frames [Yongmian and Qiang 2005].”
Facial expression recognition based on Local Binary Patterns: A comprehensive study (2009, methodology): “Dynamic Bayesian Networks (DBNs) recently were exploited for sequence-based expression recognition [16,14,18].”
Automatic facial expression recognition using features of salient facial patches (2015, methodology): “In [18], an active Infra-Red illumination along with Kalman filtering is used for accurate tracking”
U2Fusion: A Unified Unsupervised Image Fusion Network (2020, unknown): “[3], [4].”
A 3D facial expression database for facial behavior research (2006, methodology): “A number of techniques were successfully developed using 2D static images or video sequences, including machine vision techniques [44, 12, 10, 21, 4, 41, 42] and machine learning techniques [1, 20, 5, 6, 45].”
Facial Action Unit Recognition by Exploiting Their Dynamic and Semantic Relationships (2007, methodology): “The local methods [4], [9], [10], [3], [11], [12], [13], [14], [15], [16] focus on the facial features locally.”
Facial Expression Recognition via a Boosted Deep Belief Network (2014, methodology): “These features can be either hand-designed [32, 33, 34, 25, 1, 27, 28, 26, 22, 9, 4, 10, 13] or learned from training images [6, 29, 15, 16, 36, 2, 35, 30].”
Occlusion Aware Facial Expression Recognition Using CNN With Attention Mechanism (2019, background): “3) Facial Expression Dataset With Real Occlusion: Some initial efforts reported on FER in real-life with natural occlusions are limited to occlusions arising from sunglasses, medical mask [38], hands [14], [39], [40] or hair [21].”
SwinFusion: Cross-domain Long-range Learning for General Image Fusion via Swin Transformer (2022, background): “A single fused image with better scene representation and visual perception facilitates subsequent practical visual applications, such as object detection, tracking, semantic segmentation, scene understanding, etc. [7]–[9].”
Facial Expression Recognition in Image Sequences Using Geometric Deformation Features and Support Vector Machines (2007, methodology): “Feature based Approaches Facial feature detection and tracking is based on active InfraRed illumination in [6], in order to provide visual information under variable lighting and head motion.”
BP4D-Spontaneous: a high-resolution spontaneous 3D dynamic facial expression database (2014, unknown): “, anger, disgust, fear, happiness, sadness, and surprise) [3,5,13].”
Social signal processing: Survey of an emerging domain (2009, methodology): “In order to address the limitations inherent in optical flow techniques such as the accumulation of error and the sensitivity to noise, occlusion, clutter, and changes in illumination, recent efforts in automatic facial expression recognition use sequential state estimation techniques (such as Kalma”