Quadruplet Cross Similarity Network for Facial Expression Recognition
The paper introduces a novel Quadruplet Cross Similarity (QCS) Network for Facial Expression Recognition (FER) that leverages cross similarity attention to refine features by maximizing inter-class differences and minimizing intra-class differences, achieving state-of-the-art performance on several FER datasets without relying on additional landmark information or external training data.
AffectNet+: Enhancing Facial Expression Recognition with Soft-Labels for Compound Emotions
This paper introduces AffectNet+, an improved facial expression dataset utilizing "soft-labels" to represent the presence of multiple emotions in a single image, addressing limitations of traditional "hard-label" approaches in capturing the complexity of human emotions.
ARBEx: Enhancing Facial Expression Recognition Using Vision Transformer and Reliability Balancing
ARBEx is a novel framework that leverages a Vision Transformer and a reliability balancing mechanism to improve the accuracy and robustness of facial expression recognition by addressing challenges such as poor class distributions, bias, and uncertainty.
GReFEL: A Novel Approach to Facial Expression Learning Using Geometry-Aware Reliability Balancing for Improved Accuracy in Handling Bias and Imbalanced Datasets
GReFEL, a novel facial expression learning framework, leverages Vision Transformers and a geometry-aware reliability balancing module to improve accuracy and mitigate biases stemming from imbalanced datasets in facial expression recognition.
Facial Expression Recognition in Students Using an Extended Vision Long Short-Term Memory Network (xLSTM-FER)
xLSTM-FER, a novel architecture based on Extended Long Short-Term Memory (xLSTM), offers a computationally efficient and highly accurate method for recognizing student facial expressions, outperforming existing CNN and ViT-based approaches.
Adaptive Label Distribution Fusion Network for Robust Facial Expression Recognition
A dual-branch adaptive distribution fusion framework is proposed to address the ambiguity problem in facial expression recognition by mining class distributions of emotions and adaptively fusing them with label distributions of samples.
Multi-Modal Self-Supervised Learning for Facial Expression Recognition in the Wild
Employing multi-task multi-modal self-supervised learning to learn rich data representations for facial expression recognition from in-the-wild video data without requiring expensive annotations.
Emotic Masked Autoencoder with Attention Fusion for Facial Expression Recognition in ABAW Competition
MAE-FaceとFusion Attentionを統合した革新的なアプローチにより、ABAWコンペティションでの表情分類の性能向上を実証。
A3lign-DFER: Comprehensive Dynamic Affective Alignment for Facial Expression Recognition with CLIP
A3lign-DFER introduces a new paradigm for dynamic facial expression recognition, enhancing alignment and achieving state-of-the-art results.
Exploring Facial Expression Recognition through Semi-Supervised Pretraining and Temporal Modeling for ABAW Competition at CVPR2024
Addressing challenges in facial expression recognition through semi-supervised pretraining and temporal modeling for improved performance.
© 2024 by Linnk AI