A Novel Attention Fusion Approach for Joint Object Detection and Classification in Facial Expression Recognition

Facial expression recognition (FER) is a crucial task in computer vision, aiming to automatically detect and recognize individuals' emotional states by analyzing facial images. In this paper, we propose a novel attention module, the Dual-Directional Attention Mechanism (DDAM), based on an attention fusion mechanism. This module is integrated into the feature extraction network of the YOLOX framework to address joint object detection and classification tasks in unaligned images. The proposed method incorporates Mamba for contextual understanding, Neighborhood Coordinate Attention (NCA) for capturing local neighborhood relationships, Coordinate Attention (CA) for directional awareness, and Batch Attention (BA) for intrabatch relationship modeling. Experimental results demonstrate that our method outperforms mainstream FER approaches on the unaligned SFEW dataset. Furthermore, comparisons with state-of-the-art methods show that our model achieves superior performance in terms of mean Average Precision (mAP).

• Publication year

  2025

• Venue

  2025 IEEE 8th International Conference on Pattern Recognition and Artificial Intelligence (PRAI)

• Publication date

  2025-08-15

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/PRAI67447.2025.11412830
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• FER-YOLO-Mamba: Facial Expression Detection and Classification Based on Selective State Space

  2024cited by this paper
• Mamba: Linear-Time Sequence Modeling with Selective State Spaces

  2023cited by this paper
• A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS

  2023cited by this paper
• BatchFormer: Learning to Explore Sample Relationships for Robust Representation Learning

  2022cited by this paper
• YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors

  2022cited by this paper
• Neighborhood Attention Transformer

  2022cited by this paper
• YOLOX: Exceeding YOLO Series in 2021

  2021cited by this paper
• A Survey on Vision Transformer

  2020cited by this paper
• YOLOv4: Optimal Speed and Accuracy of Object Detection

  2020cited by this paper
• Objects as Points

  2019cited by this paper
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

  2019cited by this paper
• Transfer learning using VGG-16 with Deep Convolutional Neural Network for Classifying Images

  2019cited by this paper
• YOLOv3: An Incremental Improvement

  2018cited by this paper
• Attention is All you Need

  2017cited by this paper
• Focal Loss for Dense Object Detection

  2017cited by this paper
• Feature Pyramid Networks for Object Detection

  2016cited by this paper
• Resnet in Resnet: Generalizing Residual Architectures

  2016cited by this paper
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization

  2016cited by this paper
• Histograms of Oriented Gradients

  2015cited by this paper
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

  2015cited by this paper
• You Only Look Once: Unified, Real-Time Object Detection

  2015cited by this paper
• SSD: Single Shot MultiBox Detector

  2015cited by this paper
• Neural Machine Translation by Jointly Learning to Align and Translate

  2014cited by this paper
• Microsoft COCO: Common Objects in Context

  2014cited by this paper
• Categorical and dimensional affect analysis in continuous input: Current trends and future directions

  2013cited by this paper
• ImageNet classification with deep convolutional neural networks

  2012cited by this paper
• Static facial expression analysis in tough conditions: Data, evaluation protocol and benchmark

  2011cited by this paper
• Description of interest regions with local binary patterns

  2009cited by this paper

• No citing papers are available for this paper.