My name is Padam Thapa, an MS student in Computer Science at the University of New Orleans (UNO), specializing in Machine Learning and Deep Learning for Computer Vision. My research focuses on leveraging advanced AI techniques to identify and localize defects in levee systems using sensor data, UAV-captured imagery, and synthetic datasets generated through GANs, Diffusion Models, and Unity3D. I have developed robust pipelines for real-time segmentation, object detection, and annotation automation, significantly improving efficiency and accuracy in levee fault detection. My work includes building scalable full-stack pipelines for real-time inference, enhancing dataset diversity with diffusion models, and applying model quantization techniques to improve efficiency.
The ML model where the malware is detected automatically when the user happens to interact with the Trojans or Worms. Out of the 3 algorithms used, XGBoost outperformed the best.
An AI model trained with Nepal Stock Exchange dataset to optimize stock trading strategy and outcome with maximum investment return using Deep Reinforcement Learning and LSTMs.
Used highly imbalanced transactions dataset with 3 classification models trained with both imbalanced and synthetically balanced datasets.
Used a ResNet-18 model and trained it on a COVID-19 Radiography dataset to create an Image Classification model that can predict Chest X-Ray scans with reasonably high accuracy
A multi-class Classification project using Deep Convolutional Neural Network Model which can classify the images of 43 distinct types of Traffic Signals.
A Deep CNN and ResNet Model trained for classification of scenary from Satellite Images using GradCam visualization technique which helps to explain how AI models think.
Leave a response.
Or just say hi!.
United States of America