• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Development of an Interactive Tutorial on Object Recognition Based on the Yolo-v4 Neural Network Algorithm

Student: Khanaev Anton

Supervisor: Anna S. Toporkova

Faculty: HSE Tikhonov Moscow Institute of Electronics and Mathematics (MIEM HSE)

Educational Programme: Information Science and Computation Technology (Bachelor)

Final Grade: 8

Year of Graduation: 2021

Due to the significant development of machine learning and neural networks, a large number of various technologies, methods and algorithms for object detection have appeared recently. Each of them has both advantages and disadvantages. Some were designed for accurate predictions, while others were designed to work quickly on autonomous systems in real time. New algorithms are starting to come out more often, but both the quality and the amount of training material devoted to a detailed analysis of the construction of algorithmic architectures with illustrative examples is decreasing every year. Also, with the release of each new architecture or algorithm, it becomes more difficult to recognize the hidden intuitions that the authors adhered to when developing their solutions for object detection problems. This paper describes the development of an interactive tutorial for object recognition based on the existing SotA (State of the Art) method - YOLOv4. The developed material is a sequential tutorial in the form of code, images, animations, mathematical formulas and text descriptions. The results of the work are a training manual on object detection, the developed YOLOv4 algorithm, the application of the algorithm and the comparison of the results of work for different real problems. The paper consists of 52 pages, 32 sources, 36 figures and 4 tables.

Full text (added May 12, 2021)

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