MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202541015889 A) filed by Ram Kumar N; Sriram R; Sanjay A R; Kishore Nath C S; and Saranya K on February 24, 2025, for Real Time Emotional Recongnition For Enhancing Teacher Student Interaction.

Inventors include Ram Kumar N; Sriram R Number; Sanjay A R; Kishore Nath C S; and Saranya K.

The application for the patent was published on August 28, 2026, under issue no. 35/2026.

Abstract: With the rapid expansion of online education, assessing student engagement in virtual classrooms has become a crucial challenge. Traditional methods, such as survys or manual observation, are inefficient and subjective. This research proposes a deep learning-based face emotion detection system to evaluate student interest in real- time. The model leverages a hybrid CNN-LSTM architecture, where Convolutional Neural Networks (CNNs) extract spatial facial features, while Long Short-Term Memory (LSTM) networks capture temporal variations in expressions. Furthermore, Bidirectional LSTM (Bi-LSTM) is incorporated to analyze forward and backward dependencies, improving recognition accuracy. An attention mechanism is introduced to enhance focus on key facial iegions, such as eyes and mouth, where emotional cues are most significant. The system is trained on a diverse dataset of online learning scenarios to classify engagement levels, such as interest, boredom, and confusion. Experimental results show that the proposed model outperforms traditional CNN-based emotion detection by effectively capturing both spatial and temporal dynamics. ·This approach provides a real-time, automated, and scalable solution for educators to monitor student engagement, enabling adaptive and personalized online learning experiences. Keywords: Online Education, Student Engagement, Virtual Classrooms, Deep Learning, Face Emotion Detection, CNN-LSTM, Convolutional Neural Networks, Long Short-Term Memory, Bidirectional LSTM, Attention Mechanism, Facial Feature Extraction, Temporal Expression Analysis, Real-Time Monitoring, Adaptive Learning, Personalized Education.

Disclaimer: Curated by HT Syndication.