MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641080363 A) filed by Dr. R. Ganesh Babu; Dr. Y. V. Raghava Rao; Dr. S. Nizamudeen; Dhannya J; Dr. Tarun Kumar; Dr. Hanumantu Krishna Murty Dora; and Dr. Arun Kumar Sao on June 30, 2026, for Deep Learning-Based System And Method For Human Activity Recognition Using Iot-Connected Wearable Cameras.
Inventors include Dr. R. Ganesh Babu; Dr. Y. V. Raghava Rao; Dr. S. Nizamudeen; Dhannya J; Dr. Tarun Kumar; Dr. Hanumantu Krishna Murty Dora; and Dr. Arun Kumar Sao.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present invention relates to the development of a HAR is highly necessary for healthcare, security, and smart environments. Sensor-based approaches are generally deficient of necessary visual data. In this research work, a deep learning system and method have been introduced for HAR using IoT-enabled wearable cameras. In the introduced system, the first-person perspective video streams are captured and sent through the use of IoT protocol to the edge-cloud environment where processing takes place efficiently. We use a lightweight CNN network, designed particularly for egocentric vision, owing to its high dimensionality and temporal properties. Our method can efficiently extract spatial-temporal features for classifying complex human activities. We found that the performance of our method is much better than those of conventional wearable sensor-based techniques. FIG.1
Disclaimer: Curated by HT Syndication.
