CERN: Compact facial expression recognition net

Gera, Darshan; Balasubramanian, Siva; Jami, Anwesh

doi:10.1016/j.patrec.2022.01.013

Cited by 27 publications

(18 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The EfficientFace model, in contrast to the other state‐of‐the‐art methods, attains the right trade‐off between recognition accuracy and computational efficiency. Also, the compact expression recognition network (CERN) introduced by Gera et al 76 with a recognition accuracy of 62.06%, and the number of model parameters equal to 1.45M seems suitable for real‐world applications. One can deploy the neural network easily on resource‐constrained embedded devices and use it for FER in real‐time.…”

Section: Resultsmentioning

confidence: 99%

A dual‐channel ensembled deep convolutional neural network for facial expression recognition in the wild

Saurav

Saini

Singh

2023

Computational Intelligence

View full text Add to dashboard Cite

Facial expression recognition (FER) in the wild is an active and challenging field of research. A system for automatic FER finds use in a wide range of applications related to advanced human–computer interaction (HCI), human–robot interaction (HRI), human behavioral analysis, gaming and entertainment, etc. Since their inception, convolutional neural networks (CNNs) have attained state‐of‐the‐art accuracy in the facial analysis task. However, recognizing facial expressions in the wild with high confidence running on a low‐cost embedded device remains challenging. To this end, this study presents an efficient dual‐channel ensembled deep CNN (DCE‐DCNN) for FER in the wild. Initially, two DCNNs, namely the and , are trained separately on the grayscale and Scharr‐convolved vertical gradient facial images, respectively. The proposed network later integrates the two pre‐trained DCNNs to obtain the dual‐channel integrated DCNN (DCI‐DCNN). Finally, all three neural networks, namely the , , and DCI‐DCNN, are jointly fine‐tuned to get a single dual‐channel‐multi‐output model. The multi‐output model produces three prediction scores for the given input facial image. The prediction scores are thus fused using the max‐voting ensemble scheme to obtain the DCE‐DCNN with the final classification label. On the FER2013, RAF‐DB, NCAER‐S, AffectNet, and CKPlus benchmark FER datasets, the proposed DCE‐DCNN consistently outperforms the two individual DCNNs and numerous state‐of‐the‐art CNNs. Moreover, the network achieves competitive recognition accuracy on all four FER in the wild datasets with reduced memory storage size and parameters. The proposed DCE‐DCNN model with high throughput on resource‐limited embedded devices is suitable for applications that seek real‐time classification of facial expressions in the wild with high confidence.

show abstract

Section: Resultsmentioning

confidence: 99%

A dual‐channel ensembled deep convolutional neural network for facial expression recognition in the wild

Saurav

Saini

Singh

2023

Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…On the Pose-FERPlus dataset, it is increased by 5.63% and 6.17%, respectively, compared to the RAN [12] model. The advantages of the method proposed by this paper are also evident when it is compared [12] 82.72 MA-Net [13] 83.65 ASF [55] 83.95 AMP-Net [14] 85.28 CERN [44] 83.40 CT-DBN [37] 84.90 Proposed method 87.21…”

Section: Quantitative Performance Comparisonsmentioning

confidence: 91%

“…Zhao et al [13] and Liu et al [14] implemented the convolutional block attention module (CBAM) [43] module to extract locally salient features from different cropped regions. Similarly, Darshan et al [44] divided the face globally into four regions and harvested local features from these areas. This is dependent on the methodology used for facial region cropping and division, potentially impacting its performance when addressing diverse facial expressions.…”

Section: Attention Mechanism In Fermentioning

confidence: 99%

A discriminative multiscale feature extraction network for facial expression recognition in the wild

Wen,

Zhou,

Gan

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Driven by advancements in deep learning technologies, substantial progress has been achieved in the field of facial expression recognition over the past decade, while challenges remain brought about by occlusions, pose variations and subtle expression differences in unconstrained (wild) scenarios. Therefore, a novel multiscale feature extraction method is proposed in this paper, that leverages convolutional neural networks to simultaneously extract deep semantic features and shallow geometric features. Through the mechanism of channel-wise self-attention, prominent features are further extracted and compressed, preserving advantageous features for distinction and thereby reducing the impact of occlusions and pose variations on expression recognition. Meanwhile, inspired by the large cosine margin concept used in face recognition, a center cosine loss function is proposed to avoid the misclassification caused by the underlying interclass similarity and substantial intra-class feature variations in the task of expression recognition. This function is designed to enhance the classification performance of the network through making the distribution of samples within the same class more compact and that between different classes sparser. The proposed method is benchmarked against several advanced baseline models on three mainstream wild datasets and two datasets that present realistic occlusion and pose variation challenges. Accuracies of 89.63%, 61.82%, and 91.15% are achieved on RAF-DB, AffectNet and FERPlus, respectively, demonstrating the greater robustness and reliability of this method compared to the state-of-the-art alternatives in the real world.

show abstract

“…We also listed several state-of-the-art methods on the Pose-RAFDB and Pose-Affect database in recent years. The detailed recognition results are given in Table 9, from which we can find that SPA-SE network performs better than most of techniques expect Grea et al [50] and Zhao et al [52], where both global and local patches involved for facial expression recognition. In practice, owing to the distinction of feature-levels, they usually require to design different loss functions between global and local patches for improving classification accuracy.…”

Section: Experiments On Real-world Scenariosmentioning

confidence: 98%

“…As a result, a total of contain 2933 facial images are selected in Pose-AffectNet dataset, where 1,948 emotional images are collected in head pose greater 30 degree (> 30°) and 985 emotional images are collected in head pose greater 45 degree (> 45°). In this paper, we selected same experimental setting in [15,50,51,52] for pose-variant FER, where seven facial expressions (anger (AN), disgust (DI), fear (FE), happy (HA), neural (NE), sadness (SA) and surprise (SU)) and two head poses (> 30° and > 45°) were extracted, the corresponding facial images can be found in Fig. 7d.…”

Section: Datasets and Protocolsmentioning

confidence: 99%

Deep Spatial and Channel Sliding Attention Patches for Pose-invariant Facial Expression Recognition

Liu

Chen

et al. 2023

Preprint

View full text Add to dashboard Cite

Pose-invariant facial expression recognition (FER) is an import yet challenging research topics in computer vision, especially with the introduction of pose change and self-occlusion, which makes the recognition results changing from one observational angle to another. In this paper, a newly deep sliding patch combined with spatial-channel attention network is designed for pose-invariant FER. The new network architecture consists of three main components, namely, the sliding patch (SP) model, deep channel squeeze-and-extraction model (SE) and spatial sliding patch attention (SPA) model. The sliding patch (SP) model is devised to find the best patch size and stride for pose-invariant FER. The sliding patch attention (SPA) model can guide the network pay more attention to the current local patches and adaptively estimate the significance of each local sliding patch. The squeeze-and-extraction (SE) operation was implemented prior to SPA model whose aims to shrink the number of feature maps and provide more salient semantic feature maps for SPA model. Extensive experiments were implemented on four pose robust expression datasets, namely, BU3DFEP1, BU3DFEP2, Multi-PIE, Pose-RAFDB, Pose-Affect, and the experimental results indicated the validity and feasibility of SPA-SE network.

show abstract

CERN: Compact facial expression recognition net

Cited by 27 publications

References 12 publications

A dual‐channel ensembled deep convolutional neural network for facial expression recognition in the wild

A dual‐channel ensembled deep convolutional neural network for facial expression recognition in the wild

A discriminative multiscale feature extraction network for facial expression recognition in the wild

Deep Spatial and Channel Sliding Attention Patches for Pose-invariant Facial Expression Recognition

Contact Info

Product

Resources

About