Giseop Noh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Giseop Noh has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Junjun Zhang
Henan University of Engineering
3 shared publications - Fengxi Zhang
School of Software, Henan University of Engineering, Zhengzhou, China
1 shared publication - Han Pyo Lee
Hongik University
1 shared publication - Jianjing Mao
School of Software, Zhengzhou University of Industrial Technology, Zhengzhou, China
1 shared publication - Yanyang Hou
School of Information Engineering, Zhengzhou University of Industrial Technology, Zhengzhou 451100, China
1 shared publication
Latest Publications
- Open Access
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 28229-28244
SADGR: Adaptive Cross-Modal Emotion Recognition via Self-Supervised Alignment and Dynamic Gating
In this paper, we propose SADGR, an adaptive cross-modal emotion recognition framework that follows an “align first, modulate second, and fuse later” paradigm. First, we introduce a self-supervised text–audio contrastive alignment (STA-CA) stage, which leverages large-scale unlabeled audio–text data to map heterogeneous modalities into a shared semantic space prior to fusion. Second, we design an audio-guided visual gating (AG-VR) mechanism, where emotionally salient acoustic cues dynamically modulate visual representations, suppressing redundant information while enhancing emotion-relevant patterns. Extensive experiments on the CMU-MOSI and CMU-MOSEI benchmarks demonstrate that SADGR consistently outperforms state-of-the-art methods across both regression and classification tasks, achieving superior performance in ACC-7, ACC-2, and F1-score while maintaining competitive correlation results. These results indicate that explicit self-supervised alignment combined with dynamic cross-modal gating provides an effective and practical solution for robust multimodal emotion recognition.
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 28229-28244
Open AccessSADGR: Adaptive Cross-Modal Emotion Recognition via Self-Supervised Alignment and Dynamic Gating
In this paper, we propose SADGR, an adaptive cross-modal emotion recognition framework that follows an “align first, modulate second, and fuse later” paradigm. First, we introduce a self-supervised text–audio contrastive alignment (STA-CA) stage, which leverages large-scale unlabeled audio–text data to map heterogeneous modalities into a shared semantic space prior to fusion. Second, we design an audio-guided visual gating (AG-VR) mechanism, where emotionally salient acoustic cues dynamically modulate visual representations, suppressing redundant information while enhancing emotion-relevant patterns. Extensive experiments on the CMU-MOSI and CMU-MOSEI benchmarks demonstrate that SADGR consistently outperforms state-of-the-art methods across both regression and classification tasks, achieving superior performance in ACC-7, ACC-2, and F1-score while maintaining competitive correlation results. These results indicate that explicit self-supervised alignment combined with dynamic cross-modal gating provides an effective and practical solution for robust multimodal emotion recognition.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 131460-131475
Diagnosing Candidate Quality Bottlenecks in RL-Guided LLM Code Repair
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 131460-131475
Open AccessDiagnosing Candidate Quality Bottlenecks in RL-Guided LLM Code Repair
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 114242-114259
Sensor-Limited Critical-Bus Voltage Forecasting for Virtual Metering in DER-Rich Distribution Networks
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 114242-114259
Open AccessSensor-Limited Critical-Bus Voltage Forecasting for Virtual Metering in DER-Rich Distribution Networks
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- p. 1
A Metric-Aware Analysis of Trigger-Guided Adapter Training for Hallucination Mitigation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- p. 1
Open AccessA Metric-Aware Analysis of Trigger-Guided Adapter Training for Hallucination Mitigation
- Open Access
- 28 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 202340-202353
A Real-Time Operator-Centric Patch Extraction and Analysis Platform for Industrial Semiconductor Wafer C&C Processes
In semiconductor manufacturing, accurate detection of anomalies in high-speed rotating wafers is crucial for ensuring product quality and yield. Conventional methods for region of interest (ROI) selection in defect detection often rely on manual operator input, which is time-consuming and prone to inconsistency. This study presents a web-based patch extraction system capable of real-time ROI identification and monitoring with reduced manual selection. The proposed system integrates a high-speed vision module with HSV-based color space segmentation to automatically extract wafer surface patches while enabling operator-centric interaction through an intuitive web interface. Additionally, a Vision Transformer (ViT)-based backend was implemented and evaluated to assess its anomaly detection performance on the extracted patches. Experimental results demonstrate that the proposed method improves detection accuracy and processing speed compared to manual ROI selection, enabling efficient and consistent monitoring in wafer inspection processes. The system’s modular architecture also facilitates future integration with advanced deep learning models and deployment in diverse manufacturing environments.
- 28 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 202340-202353
Open AccessA Real-Time Operator-Centric Patch Extraction and Analysis Platform for Industrial Semiconductor Wafer C&C Processes
In semiconductor manufacturing, accurate detection of anomalies in high-speed rotating wafers is crucial for ensuring product quality and yield. Conventional methods for region of interest (ROI) selection in defect detection often rely on manual operator input, which is time-consuming and prone to inconsistency. This study presents a web-based patch extraction system capable of real-time ROI identification and monitoring with reduced manual selection. The proposed system integrates a high-speed vision module with HSV-based color space segmentation to automatically extract wafer surface patches while enabling operator-centric interaction through an intuitive web interface. Additionally, a Vision Transformer (ViT)-based backend was implemented and evaluated to assess its anomaly detection performance on the extracted patches. Experimental results demonstrate that the proposed method improves detection accuracy and processing speed compared to manual ROI selection, enabling efficient and consistent monitoring in wafer inspection processes. The system’s modular architecture also facilitates future integration with advanced deep learning models and deployment in diverse manufacturing environments.