The goal of this work was to create a web interface for marking image sequences for the purpose of collecting data needed for deep learning. The web application should offer the option of user registration, login and logout, while differentiating user roles. The administrator has control over all user accounts, editing page posts as well as all instances of classes in the application. The latter includes adding, deleting or modifying images and any masks created from them. The role of the user is much smaller, the possibility of logging in, registering or logging out, and the possibility of viewing and marking image elements. The work is conceptually divided into three parts. The first is a theoretical introduction to all the technologies used in creating a web application, starting from the server side to the client side. Then follows a detailed description of how the application was made step by step and how it all works together. The third part is documentation on how to use the web application and is intended for administrators as well as ordinary users. In order to realize the above, we used the Django framework and its integrated SQL Lite database together with other web technologies such as HTML, CSS, JavaScript and Ajax. In the practical part of the work, the process of setting up the database, server side and client side is described first. Then the implementation of various functionalities, methods and components of the application is described. The mode of operation of the application through its use is also shown, showing all the interfaces that the user may encounter during use.