Platform for Finding Similar Sneakers

Student: Vilkov Timofey

Educational Programme: Machine Learning and Data-Intensive Systems (Master)

Year of Graduation: 2024

In this work, the task of implementing a platform for recommending the most relevant objects from a database based on a user-written text query using multimodal models is addressed. The result of the work is a service that allows users to obtain the most relevant pairs of shoes for their query through a visual interface. Additionally, functionality is available for filtering the set on which the search is performed, for example, by price. The Qdrant vector database is used for storing information. To accelerate the service, a mechanism for semantic caching is implemented. Using the fashion domain as an example, the work demonstrates that combining information from multiple modalities allows the model to achieve quality comparable to fine-tuning the original model for a specific domain, using only the information extracted from the image

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