Action-Recognition-Soccer-Players

A GUI which enables action classification and labelling of soccer player’s action in a broadcast video based on teams. Focus Areas: dribbling,kicking, running and passing using over 2000 sample images Methodology Create a flask application Connect to the User Interface(React) Use the trained model to process video frames Detect the skeleton(s) in each frame Label each detected skeleton(s) Give live analytics from the processing based on each team. Display the processed frames in a video format for the user. Get real-time action metrics for each team.

Soccer-Player-Action-Recognition-Application

A GUI which enables action classification and labelling of soccer player’s action in a broadcast video based on teams. Focus Areas: dribbling,kicking, running and passing using over 2000 sample images Methodology Create a flask application Connect to the User Interface(React) Use the trained model to process video frames Detect the skeleton(s) in each frame Label each detected skeleton(s) Give live analytics from the processing based on each team. Display the processed frames in a video format for the user. Get real-time action metrics for each team.

Soccer-Players-Detection-Detectron2

This script is used for object detection and tracking of soccer players with background color segmentation of the soccer pitch using OpenCV in Detectron2. It is still a work in progress and the end goal is to get a model that can fully classify players on the field according to their teams with this process.