No abstract

No abstract

In this paper, we address the problem of spatio-temporal person retrieval from multiple videos using a natural language query, in which we output a tube (i.e., a sequence of bounding boxes) which encloses the person described by the query. For this problem, we introduce a novel dataset consisting of videos containing people annotated with bounding boxes for each second and with five natural language descriptions. To retrieve the tube of the person described by a given natural language query, we design a model that combines methods for spatio-temporal human detection and multimodal retrieval. We conduct comprehensive experiments to compare a variety of tube and text representations and multimodal retrieval methods, and present a strong baseline in this task as well as demonstrate the efficacy of our tube representation and multimodal feature embedding technique. Finally, we demonstrate the versatility of our model by applying it to two other important tasks.We will make the code, features, and annotations pub-

Thesis is devoted to the creation of computer animation for character's 3Dmodel on the base of motion capture technology.The first section describes the concept of computer animation, its types and applications. The basics of motion capture technology as well as a comparative analysis of different types of motion capture systems, their advantages and disadvantages. The process of creating face animation based on motion capture technology is also considered.The second section provides a rationale for the choice of tools and working material to perform the practical part of the work.The third section presents the practical side of the process of creating computer animation and analyzes the results. The analysis showed the presence of some shortcomings both in the work of means of creating animation based on motion capture technology, and the technology as a whole.