Publications

2025

1. Unleashing the Power of Dynamic Mode Decomposition and Deep Learning for Rainfall Prediction in North-East India

   Paleti Nikhil Chowdary, Sathvika Pingali, Pranav Unnikrishnan, and 4 more authors

   In Computation of Artificial Intelligence and Machine Learning, 2025

   Abs

   Accurate rainfall forecasting is crucial for effective disaster preparedness and mitigation in the North-East region of India, which is prone to extreme weather events such as floods and landslides. In this study, we investigated the use of two data-driven methods, dynamic mode decomposition (DMD) and long-short-term memory (LSTM), for rainfall forecasting using daily rainfall data collected from the India Meteorological Department in northeast region over a period of 122 years. We conducted a comparative analysis of these methods to determine their relative effectiveness in predicting rainfall patterns. Using historical rainfall data from multiple weather stations, we trained and validated our models to forecast future rainfall patterns. Our results indicate that both DMD and LSTM are effective in forecasting rainfall, with LSTM outperforming DMD in terms of accuracy, revealing that LSTM has the ability to capture complex nonlinear relationships in the data, making it a powerful tool for rainfall forecasting. The study reveals that the DMD method achieved Mean Squared Error (MSE) values ranging from 150.44 mm to 263.34 mm and Mean Absolute Error (MAE) values from 91.34 mm to 154.61 mm. In contrast, the Deep Learning (DL) approach, utilizing LSTM, demonstrated a normalized MAE value of 0.35 and a normalized RMSE value of 0.534. Our findings suggest that data-driven methods such as DMD and deep learning approaches like LSTM can significantly improve rainfall forecasting accuracy in the North-East region of India, helping to mitigate the impact of extreme weather events and enhance the region’s resilience to climate change.

2024

1. Comparative Analysis on Speech Driven Gesture Generation

   Pranav Unnikrishnan, K. S. R. Logesh, Abinesh Sivakumar, and 3 more authors

   In Artificial Intelligence and Knowledge Processing, 2024

   Abs

   This study presents a comparative analysis of various deep- learning methods for gesture generation from speech, contributing to the field of human-agent interaction. The primary focus lies in advancing interactions with virtual agents and robots through the application of varied representation learning approaches. The methodologies employ Gated Recurrent Unit (GRU), Bidirectional-GRU, and Multi-head attention techniques to build a concise representation of human gestures, acting as motion encoders and decoders. Incorporating a pre-existing gesture generation approach, the established groundwork through the utilization of a network named Speech E is leveraged. Furthermore, an in-depth analysis of diverse speech feature inputs’ impact on the model’s performance is pursued. This network is designed to efficiently transform speech input into a gesture representation with reduced dimensions. The influence of different speech feature inputs on the model’s performance is explored. The integration of GRU, Bidirectional-GRU, and Multi-Head Attention methods allow a thorough evaluation of their effectiveness in translating speech into corresponding gestures. This study provides insights into the potential of these techniques, and their implications for creating more intuitive and responsive virtual agents and robots. Notably, our investigations exclusively utilize Mel-Frequency Cepstral Coefficients (MFCCs) as features, revealing the optimal performance of our models. This careful feature selection shines brightly in our results, where our Bi-directional GRU model outshines the rest.

2023

1. Face Mask Detection Using Transfer Learning and TensorRT Optimization

   Paleti Nikhil Chowdary, Pranav Unnikrishnan, Rohan Sanjeev, and 4 more authors

   In International Conference on Innovative Computing and Communications, 2023

   Abs

   TensorRT is a high-performance deep learning inference optimizer and runtime that can be used to speed up the deployment of deep learning models. In this paper, the performance of different neural network architectures when using TensorRT is compared and showed that TensorRT can significantly reduce the inference time of deep learning models on embedded systems. The SARS-CoV-2 virus, which causes the infectious disease COVID-19, has had a significant impact on global health and the economy. Non-pharmaceutical interventions such as wearing face masks are an effective way to reduce the spread of the virus, and automatic detection systems based on CNN’s can help to detect mask-wearing without requiring human intervention which saves resources or manpower deployed. The results demonstrate that TensorRT is a valuable tool for deploying deep learning applications in resource-constrained environments and can help to improve the performance of a wide range of neural network architectures.