• Conditional MedicalGAN for Data Augmentation - [Github]

This project enerates images based on a provided label or condition while utilizing Deep Convolutional Neural Networks. This approach is particularly useful in addressing class imbalance issues in image datasets. My experience with this technique has been primarily in 2 datasets, namely the Pneumonia and Covid Lung CT datasets.

• 3D Object Recognition - [Github]

This project aimed for 3D Object Recognition using the LiDAR sensor an improvement on Complex-YOLO, a real-time 3D object detection network designed exclusively for point clouds fine-tuned by the YOLO-NAS model.

• Providing Thermal Comfort at Home/Hospitals - [Github]

This Project assesses thermal comfort using algorithms that predict temperature, humidity, and air velocity with SVR and classify thermal sensation levels with semi-supervised algorithms. It is scalable and adaptable, providing a responsive and nuanced assessment of thermal comfort.