Vincentqyw/image-matching-webui
A Gradio-based web interface for running multiple deep learning-based image matching algorithms on image pairs or webcam feeds.

Image Matching WebUI (IMCUI) provides a graphical interface to apply various computer vision algorithms for matching keypoints between two images. The project integrates several state-of-the-art deep learning models including SuperPoint, SuperGlue, LightGlue, LoFTR, TopicFM, and ASpanFormer for feature detection, description, and matching. Users can select images from local files or a webcam, choose an algorithm, and visualize the matching results with overlaid keypoints and correspondences.