Facial Expression Recognition Methods in the Wild Based on Fusion Feature of Attention Mechanism and LBP

Abstract

Facial expression methods play a vital role in human–computer interaction and other fields, but there are factors such as occlusion, illumination, and pose changes in wild facial recognition, as well as category imbalances between different datasets, that result in large variations in recognition rates and low accuracy rates for different categories of facial expression datasets. This study introduces RCL-Net, a method of recognizing wild facial expressions that is based on an attention mechanism and LBP feature fusion. The structure consists of two main branches, namely the ResNet-CBAM residual attention branch and the local binary feature (LBP) extraction branch (RCL-Net). First, by merging the residual network and hybrid attention mechanism, the residual attention network is presented to emphasize the local detail feature information of facial expressions; the significant characteristics of facial expressions are retrieved from both channel and spatial dimensions to build the residual attention classification model. Second, we present a locally improved residual network attention model. LBP features are introduced into the facial expression feature extraction stage in order to extract texture information on expression photographs in order to emphasize facial feature information and enhance the recognition accuracy of the model. Lastly, experimental validation is performed using the FER2013, FERPLUS, CK+, and RAF-DB datasets, and the experimental results demonstrate that the proposed method has superior generalization capability and robustness in the laboratory-controlled environment and field environment compared to the most recent experimental methods.

A Multi Feature Fusion Search for Facial Expression Recognition Based on Classical Extraction Methods

“…The authors infer that the two main branches that make up the structure are LBP extraction branch and residual attention branch for ResNet-CBAM (Liao, et al, 2023). To this end, in building the classification model for the residual attention, the discriminative features of the face expressions are extracted from both the channel and spatial dimensions, the residual attention is shown to emphasize the local features are categorized as expressions (Liao, et al, 2023). This approach seek to combine the attention mechanism and the residual network.…”

“…The results obtained from the experiments demonstrates how effective the proposed approach outperforms as compared to recent experimental results and this is evident in the robustness and the generalization of both laboratory-controlled environments and field environments. (Liao, et al, 2023) 2.6 Distinctive Image Features from Scale-Invariant Keypoints Lowe (2004) provides image features for matching various photographic based image content retrieved from various object attributes. The rotation and scaling characteristics of the images with the lighting changes and 3D angle position of the camera are discriminative characters, for this reason the localized frequency and spatial are excellent in that they are resistant to noise, occluded or cluttered (Lowe, 2004).…”

“…In this study, the authors Liao, et al (2023) introduce recognition for facial expression using Wild based on feature fusion of LBP and attention mechanism. The authors infer that the two main branches that make up the structure are LBP extraction branch and residual attention branch for ResNet-CBAM (Liao, et al, 2023).…”

“…In this study, the authors Liao, et al (2023) introduce recognition for facial expression using Wild based on feature fusion of LBP and attention mechanism. The authors infer that the two main branches that make up the structure are LBP extraction branch and residual attention branch for ResNet-CBAM (Liao, et al, 2023). To this end, in building the classification model for the residual attention, the discriminative features of the face expressions are extracted from both the channel and spatial dimensions, the residual attention is shown to emphasize the local features are categorized as expressions (Liao, et al, 2023).…”

Comparison and Analysis of CNN Models to Improve a Facial Emotion Classification Accuracy for Koreans and East Asians

“…However, in noisy (wild) environments, they generally do not work well due to changes in head posture, lighting, etc. In recent research, attention mechanisms for image classification problems have been developed to increase the performance of CNN by focusing on small details [5]. Moreover, in image segmentation problems, CNN effectively derives valid data by searching pixel units in images and classifying them into practical semantic units [6].…”