Easy Zoom In: Super-Resolution of Cityscapes
Project|2023
UI
Easy Zoom In: Super-Resolution of Cityscapes is a system developed as a final project for the "Computer Vision Practice with Deep Learning (CVPDL)" course at National Taiwan University (Spring 2023). It aims to enhance the visual quality of low-resolution dashcam or monitor video feeds, particularly focusing on improving the details of objects like cars, pedestrians, and riders. This is achieved by combining YOLOv7 for object detection with Latent Diffusion Models (LDM) for super-resolution of the detected and cropped regions, offering a cost-effective alternative to expensive hardware upgrades.