s mohammadi

As a Senior Data Scientist, I specialize in computer vision applications for real estate, focusing on virtual staging, image enhancement, and 3D object reconstruction. I leverage generative AI and segmentation techniques to deliver innovative solutions.

In my role as a Data Scientist, I focused on extracting personal information from national identification cards and developed image datasets. I fine-tuned OCR models and implemented voice assistant technologies.

As a Deep Learning Developer, I implemented face recognition systems and developed APIs for hand gesture recognition and COVID-19 diagnosis from CT scans. I utilized various deep learning frameworks for model training and deployment.

Thesis: Brain tissue segmentation in the presence of MS lesions and INU artifact. The proposed method is an end-to-end method that is a combination of adversarial generative networks and a segmentation module to achieve better performance in brain tissue segmentation without any additional pre-processing to remove noise and artifacts.

Thesis: Hand Gesture Recognition. Collected dataset that included 8000 images in static hand gesture classes in complex background via different volunteers. Fine-tuned ImageNet models such as VGG16, ResNet, Xception-v3, etc and Trained object detections (YOLOv4, SSD MobileNet and EfficientDet) on diverse hand gesture dataset.