test
Search publications, data, projects and authors

Thesis

Spanish

ID: <

http://hdl.handle.net/10251/63282

>

·

DOI: <

10.4995/thesis/10251/63282

>

Where these data come from
Generation of real-time synthetic image based on digitised sound samples in the digital art field

Abstract

[EN] The present research is framed within the field of sound visualization, which within the context of fine art examines the synthesis between image and sound, the physical and graphical representations of sound and its evolution in contemporary art. When sound is structured with an artistic purpose, as in music, in addition to physical or psycho-acoustic properties such as intensity and frequency, we perceive aesthetic and conceptual qualities. In this case, a visual interpretation should at least gather some of those physical, aesthetic and conceptual qualities, whilst the merger process should allow for innovative ideas. In this context, the research focuses on real-time image production drawn from the sound; proposing a sound visualization method based on numerical samples obtained as a result of digitisation. This method is designed for an environment in which music and image production are developed in real time, and are perceived jointly, being suitable for audiovisual live events. If we consider that a digitised sound is nothing but a set of numbers arranged in time, called samples, and an image is nothing more than a collection of numbers arranged in space, so-called pixeles, the method we propose takes the numerical values of the sound samples and then spatially locates them to create the pixels of an image. Over time it will generate a sequence of images. Depending on the employed strategy of the spatial placement of the samples, the number of samples (past and present) and the number of sound sources to be embedded, we obtain different families of visualization algorithms with their own aesthetic properties. The objective of our display method is to uncover visual interpretations that developing in real time allow to collect part of the physical, aesthetic and conceptual properties of sound, under the hypothesis that the internal structure of the sound signal contains perceptual properties that are independent of the sensory channel. In order to achieve this objective and seeking to corroborate the hypothesis, we have developed software to create images from sound samples and we have analyzed the results in the field of digital art. The proposed display method is based on the fact that from the data-processing point of view sound and moving images are sequences of numbers and therefore we simply suggest creating the images in accordance with the numbers of the sound. Through the design of case studies we have experimented with different sound compositions and with different strategies of spatial placement of the sound samples. After the analysis we have been able to conclude that even under the conditions of a very simple algorithms placement a close correlation between sound and image can be appreciated. When the sound lines divide into colour channels or images, the instruments can be clearly seen. The more periodical is the sound the more structured and organized is the resulting image, even with placement algorithms that do not strictly maintain the sequential order of the sound samples. In general, the aesthetic result can be framed within the context of abstract art, confirming the essentially abstract nature of sound. Future work will consider the possibility of expanding the subject of algorithms of spatial location, providing them with artificial intelligence, as well as of the further study of hybridization of these algorithms, the possibility of incorporating transformed maths based on the frequency domain and the extension of the investigation into more general data visualization. TESIS

Your Feedback

Please give us your feedback and help us make GoTriple better.
Fill in our satisfaction questionnaire and tell us what you like about GoTriple!