The work of surf photographers involves capturing images from sessions on random beaches, often resulting in many photos per session. These photos are then uploaded to web platforms such as SurfMappers, where surfers need to review all the images to find photos of themselves. This process is extremely time-consuming and inefficient, as it requires each surfer to analyze a large bank of images manually.
This repository explores solutions to solve this problem this repository extracts images from one of these websites and applies computer vision techniques to separate each surfer into a different cluster.
All this work was developed as the final project of the Computer Vision course at the Federal University of Santa Catarina. To facilitate the access of the data used, it is all available here.