TIAGO FERNANDES TAVARES
Lattes
Google ScholarUnidades Organizacionais
Resumo profissional
person.page.researcharea
Nome para créditos
8 resultados
Resultados da Pesquisa
Agora exibindo 1 - 8 de 8
Artigo Científico A multi-sensor human gait dataset captured through an optical system and inertial measurement units(2022) Santos, Geise; Wanderley, Marcelo; TIAGO FERNANDES TAVARES; Rocha, AndersonDiferent technologies can acquire data for gait analysis, such as optical systems and inertial measurement units (IMUs). Each technology has its drawbacks and advantages, ftting best to particular applications. The presented multi-sensor human gait dataset comprises synchronized inertial and optical motion data from 25 participants free of lower-limb injuries, aged between 18 and 47 years. A smartphone and a custom micro-controlled device with an IMU were attached to one of the participant’s legs to capture accelerometer and gyroscope data, and 42 refexive markers were taped over the whole body to record three-dimensional trajectories. The trajectories and inertial measurements were simultaneously recorded and synchronized. Participants were instructed to walk on a straight-level walkway at their normal pace. Ten trials for each participant were recorded and pre processed in each of two sessions, performed on diferent days. This dataset supports the comparison of gait parameters and properties of inertial and optical capture systems, whereas allows the study of gait characteristics specifc for each system.Trabalho de Evento Comparative Latency Analysis of Optical and Inertial Motion Capture Systems for Gestural Analysis and Musical Performance(2021) Santos, Geise; Wang, Johnt; Brum, Carolina; Wanderley, Marcelo M.; Tavares, Tiago; TIAGO FERNANDES TAVARES; Rocha, AndersonWireless sensor-based technologies are becoming increasingly accessible and widely explored in interactive musical performance due to their ubiquity and low-cost, which brings the necessity of understanding the capabilities and limitations of these sensors. This is usually approached by using a reference system, such as an optical motion capture system, to assess the signals’ properties. However, this process raises the issue of synchronizing the signal and the reference data streams, as each sensor is subject to different latency, time drift, reference clocks and initialization timings. This paper presents an empirical quantification of the latency communication stages in a setup consisting of a Qualisys optical motion capture (mocap) system and a wireless microcontroller-based sensor device. We performed event-to-end tests on the critical components of the hybrid setup to determine the synchronization suitability. Overall, further synchronization is viable because of the near individual average latencies of around 25ms for both the mocap system and the wireless sensor interface.Trabalho de Evento InFracta: the Body as an Augmented Instrument in a Collaborative, Multi-Modal Piece(2021) Pessanha, Thales Roel P.; Roque, Thiago Rossi; Zanchetta, Guilherme; Pereira, Lucas B.; Oliveira, Gabrielly Lima de; Pinheiro, Bruna C. M.; Paulino, Renata F. P. P.; TIAGO FERNANDES TAVARESThis paper discusses the creative process of the piece “InFracta: Dialogue Processes in a Multi-modal Environment”. The discussion concerns the dialogues between the dance, music, image, and technology knowledge domains, which were all present in the construction of the piece’s poetics. The interaction between these domains fostered resignifying the dancers’ gestures, so that their bodies interacted with a sound environment as if they were augmented instruments. This discussion adds to previous work on technology-mediated multi-modal art, in special concerning the contribution and the emergence of meaning related to each of the knowledge domains involved in the piece.Trabalho de Evento Unsupervised Improvement of Audio-Text Cross-Modal Representations(2023) Wang, Zhepei; Subakan, Cem; Subramani, Krishna; Wu, Junkai; TIAGO FERNANDES TAVARES; FABIO JOSE AYRES; Smaragdis, ParisRecent advances in using language models to obtain cross-modal audio-text representations have overcome the limitations of conventional training approaches that use predefined labels. This has allowed the community to make progress in tasks like zero-shot classification, which would otherwise not be possible. However, learning such representations requires a large amount of human-annotated audio-text pairs. In this paper, we study unsupervised approaches to improve the learning framework of such representations with unpaired text and audio. We explore domain-unspecific and domain-specific curation methods to create audio-text pairs that we use to further improve the model. We also show that when domain-specific curation is used in conjunction with a soft-labeled contrastive loss, we are able to obtain significant improvement in terms of zero-shot classification performance on downstream sound event classification or acoustic scene classification tasks.Artigo Científico Reliability and generalization of gait biometrics using 3D inertial sensor data and 3D optical system trajectories(2022) Santos, Geise; TIAGO FERNANDES TAVARES; Rocha, AndersonParticularities in the individuals’ style of walking have been explored for at least three decades as a biometric trait, empowering the automatic gait recognition feld. Whereas gait recognition works usually focus on improving end-to-end performance measures, this work aims at understanding which individuals’ traces are more relevant to improve subjects’ separability. For such, a manifold projection technique and a multi-sensor gait dataset were adopted to investigate the impact of each data source characteristics on this separability. Assessments have shown it is hard to distinguish individuals based only on their walking patterns in a subject-based identifcation scenario. In this setup, the subjects’ separability is more related to their physical characteristics than their movements related to gait cycles and biomechanical events. However, this study’s results also points to the feasibility of learning identity characteristics from individuals’ walking patterns learned from similarities and diferences between subjects in a verifcation setup. The explorations concluded that periodic components occurring in frequencies between 6 and 10 Hz are more signifcant for learning these patterns than events and other biomechanical movements related to the gait cycle, as usually explored in the literature.Trabalho de Evento Explorando o Gap de Gênero em Matemática no ENEM em Municípios Brasileiros como Indicativo de Potenciais Candidatas para Carreiras em Computação(2024) TIAGO FERNANDES TAVARES; Miranda, Fábio R.Nas ciências exatas e tecnológicas tipicamente há uma grande diferença entre o tamanho das populações masculina e feminina, o que é conhecido como gap de gênero. Esse fenômeno é observado mundialmente em disciplinas do campo STEM (Science, Technology, Engineering, and Math; ciência, tecnologia, engenharia e matemática), e é relacionado a aspectos culturais que são mensuráveis a partir do ensino médio. Neste artigo, estimamos o gap de gênero à partir da análise das notas de matemática de alunos no ENEM. A análise dos dados mostra que há cidades que consistentemente têm um gap de gênero pequeno, o que indica possíveis fontes de inspiração para ações futuras na redução deste fenômeno.Trabalho de Evento Continuous Parameter Control Using an On/Off Sensor in the Augmented Handheld Triangle(2021) Ferreira, Marcio Albano H.; TIAGO FERNANDES TAVARESIn this work, we present Triaume, an augmented musical percussion instrument based on the triangle. The augmentation proposal for this instrument is based on a capacitive thumb sensor, that allows controlling digital musical devices and at the same time, preserves the original instrument idiomatic inside the context in which it is inserted. Triaume’s interaction proposals were built upon Brazilian music genres idiomatic, such as Forro, Xote, and Bai ´ ao. The instrument ˜ invasiveness is further reduced through the use of an external device (an appli cation running on a smartphone) for emulating faders related to sound parameter configuration. At first, we used the sensor as an on/off button able to trigger pre programmed percussion samples, which can be synchronized with the triangle’s acoustic sounds. This mechanism can be adjusted for triggering on pressing or releasing the sensor. Next, we convert the digital signals acquired by the sensor to continuous values by the on/off signal filtering. Such signal rectification and filtering system allows gradual change of continuous sound synthesis parameters, dealing with the on/off signal low robustness, turning it into a more controlled sig nal. Triaume can be inserted in the contexts of traditional and avant-garde music, also motivating further studies for applying the mechanism used on it in other percussion instruments.Artigo Científico Audiovisual interactive artwork via web-deployed software: Motus composes Homino-idea(2022) Amstalden, Augusto Piato; TIAGO FERNANDES TAVARES; Costa Neto, Anésio Azevedo; Camarini, Giovana CardiMany art installations rely on camera-based audiovisual interactions, and this commonly requires specialized hardware and software. Consequently, audiovisual installations are usually restricted to wealthier areas, in which the specialized equipment can be afforded and properly hosted. In countries with an evident income unbalance linked to location, the geographic restriction leads to an audience restriction. In this work, we present the development of a web-deployed composition tool for audiovisual interactions that runs on the client side and does not require installing any additional software. Simultaneously, it provides visual feedback that can aid the audience to understand the experience. Consequently, the tool can be used to compose audiovisual interactions that reach a large audience via web. We further explore the tool by composing the audiovisual installation Homino-idea. The installation is inspired by the interactions between humans and the environment, and can be either shown in art venues or used online.