no code implementations • 22 Feb 2024 • Lala Shakti Swarup Ray, Bo Zhou, Sungho Suh, Lars Krupp, Vitor Fortes Rey, Paul Lukowicz
We show that the combination of vector quantization of sensor data along with simple text conditioned auto regressive strategy allows us to obtain high-quality generated pressure sequences from textual descriptions with the help of discrete latent correlation between text and pressure maps.
no code implementations • 31 Jan 2024 • Mengxi Liu, Vitor Fortes Rey, Yu Zhang, Lala Shakti Swarup Ray, Bo Zhou, Paul Lukowicz
While IMUs are currently the prominent fitness tracking modality, through iMove, we show bio-impedence can help improve IMU-based fitness tracking through sensor fusion and contrastive learning. To evaluate our methods, we conducted an experiment including six upper body fitness activities performed by ten subjects over five days to collect synchronized data from bio-impedance across two wrists and IMU on the left wrist. The contrastive learning framework uses the two modalities to train a better IMU-only classification model, where bio-impedance is only required at the training phase, by which the average Macro F1 score with the input of a single IMU was improved by 3. 22 \% reaching 84. 71 \% compared to the 81. 49 \% of the IMU baseline model.
no code implementations • 21 Nov 2023 • Dominique Nshimyimana, Vitor Fortes Rey, Paul Lukowic
Machine learning algorithms are improving rapidly, but annotating training data remains a bottleneck for many applications.
no code implementations • 7 Aug 2023 • Sungho Suh, Vitor Fortes Rey, Sizhen Bian, Yu-Chi Huang, Jože M. Rožanec, Hooman Tavakoli Ghinani, Bo Zhou, Paul Lukowicz
This paper presents a novel wearable sensing prototype that combines IMU and body capacitance sensing modules to recognize worker activities in the manufacturing line.
no code implementations • 1 Aug 2023 • Lala Shakti Swarup Ray, Vitor Fortes Rey, Bo Zhou, Sungho Suh, Paul Lukowicz
We propose PressureTransferNet, a novel method for Human Activity Recognition (HAR) using ground pressure information.
no code implementations • 3 Jul 2023 • Vitor Fortes Rey, Dominique Nshimyimana, Paul Lukowicz
Recently self-supervised learning has been proposed in the field of human activity recognition as a solution to the labelled data availability problem.
no code implementations • 30 May 2023 • Si Zuo, Vitor Fortes Rey, Sungho Suh, Stephan Sigg, Paul Lukowicz
The proposed method aims to generate synthetic time-series sensor data without relying on labeled data, addressing the scarcity and annotation difficulties associated with real-world sensor data.
Ranked #1 on Human Activity Recognition on MM-Fit
Generative Adversarial Network Human Activity Recognition +1
no code implementations • 22 May 2023 • Mengxi Liu, Bo Zhou, Zimin Zhao, Hyeonseok Hong, Hyun Kim, Sungho Suh, Vitor Fortes Rey, Paul Lukowicz
In this work, we propose an open-source scalable end-to-end RTL framework FieldHAR, for complex human activity recognition (HAR) from heterogeneous sensors using artificial neural networks (ANN) optimized for FPGA or ASIC integration.
1 code implementation • 26 Oct 2022 • Sizhen Bian, Vitor Fortes Rey, Siyu Yuan, Paul Lukowicz
In the second case, we tried to recognize actions related to manipulating objects and physical collaboration between users by using a wrist-worn HBC sensing unit.
no code implementations • 4 Oct 2022 • Vitor Fortes Rey, Sungho Suh, Paul Lukowicz
To mitigate this problem we propose a method that facilitates the use of information from sensors that are only present during the training process and are unavailable during the later use of the system.
no code implementations • 14 Sep 2022 • Sungho Suh, Vitor Fortes Rey, Paul Lukowicz
In this work, we propose a novel Transformer-based Adversarial learning framework for human activity recognition using wearable sensors via Self-KnowledgE Distillation (TASKED), that accounts for individual sensor orientations and spatial and temporal features.
no code implementations • 23 Oct 2021 • Sungho Suh, Vitor Fortes Rey, Paul Lukowicz
The proposed network is based on the adversarial encoder-decoder structure with the MMD realign the data distribution over multiple subjects.
no code implementations • 23 Nov 2020 • Vitor Fortes Rey, Kamalveer Kaur Garewal, Paul Lukowicz
Furthermore we show that by either including a small amount of real sensor data for model calibration or simply leveraging the fact that (in general) we can easily generate much more simulated data from video than we can collect in terms of real sensor data the advantage of real sensor data can be eventually equalized.
no code implementations • ICLR 2018 • Sugandha Doda, Vitor Fortes Rey, Dr. Nadereh Hatami, Prof. Dr. Paul Lukowicz
With LeNet-5 on MNIST, pruning reduces the number of parameters by 15. 3 times and storage requirement from 1. 7 MB to 0. 006 MB with accuracy loss of 0. 03%.