MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122239 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 4, 2025, for 'lightweight residual cnn architecture for facial expression recognition.'
Inventor(s) include Dr. Krishnaraj Natarajan; Ms. Manvendra Singh; and Mr. Kshitiz Bhargava.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure provides a facial expression recognition system comprising a lightweight residual convolutional neural network architecture configured to process facial images having dimensions of 96 96 3 pixels. The architecture includes a plurality of convolutional layers arranged in residual blocks, each residual block comprising convolutional layers with batch normalization and LeakyReLU activation functions, spatial dropout layers configured to prevent overfitting, and skip connections enabling gradient flow through deeper network layers. The system further includes a global average pooling layer configured to reduce spatial dimensions while preserving feature information, and a dense output layer with softmax activation configured to classify facial expressions into eight discrete emotion categories including surprise, fear, neutral, sadness, disgust, contempt, happiness, and anger. The network is trained from scratch without relying on pretrained models using categorical cross-entropy loss function and Adam optimizer."
Disclaimer: Curated by HT Syndication.
