Vaibhav Singh
In the rapidly evolving field of deep learning, the challenge often lies not just in designing powerful models but also in making them accessible and efficient for practical use, especially
Explore medical image segmentation using the UW-Madison dataset, fine-tune Segformer with PyTorch & HuggingFace transformers, and deploy a Gradio inference app.
Medical diagnostics rely on quick, precise image classification. Using PyTorch & Lightning, we fine-tune EfficientNetv2 for medical multi-label classification.
YOLO-NAS is the new real-time SOTA object detection model. YOLO-NAS models outperform YOLOv7, YOLOv8 & YOLOv6 3.0 models in terms of mAP and inference latency.
An in-depth explanation of the theory and math behind denoising diffusion probabilistic models (DDPMs) and implementing them from scratch in PyTorch.
In this article, we explore the Diffusion models for Image generation and art generation. We cover models like Dall-E 2, Imagen, Stable Diffusion, and Midjourney
DeepLab models, first debuted in ICLR ‘14, are a series of deep learning architectures designed to tackle the problem of semantic segmentation. After making iterative refinements through the years, the
YOLOR, inspired by how humans combine knowledge, is an object detection model that pushes the boundaries of real-time detection with improved speed & accuracy.
Driver drowsiness detection systems help reduce mishaps due to tired or sleepy drivers. Learn to build such a robust system using MediaPipe in Python.
Moving away from traditional document scanners, learn how to create a Deep Learning-based Document Segmentation model using DeepLabv3 architecture in PyTorch.
