Human Action Recognition For Boxing Training Simulatior

Student: Broilovskiy Anton

Supervisor: Ilya Makarov

Educational Programme: Data Science (Master)

Year of Graduation: 2020

Computer vision technologies are widely used in sports to control the quality of training. In this paper, we have considered how to use the technology to recognize the punches of a person engaged in Boxing training. All existing approaches are based on manual feature selection and aimed to for action classification on trimmed video. We introduce a new approach for highlighting actions in an untrimmed video based on 3 stages: removing frames that do not contain actions, action localization, and action classification. Furthermore, we collected a dataset for training and testing models that contains more than 1000 examples and 5 different classes. At each stage, we compared existing approaches and found the best solution that allowed us to recognize actions on video with accuracy 87%.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses