Search Results for author: Sungho Suh

Found 37 papers, 6 papers with code

BeSound: Bluetooth-Based Position Estimation Enhancing with Cross-Modality Distillation

no code implementations24 Apr 2024 Hymalai Bello, Sungho Suh, Bo Zhou, Paul Lukowicz

Once the student model is trained, the model only takes as inputs the BLE-RSSI data for inference, retaining the advantages of ubiquity and low cost of BLE RSSI.

Knowledge Distillation Position

Text me the data: Generating Ground Pressure Sequence from Textual Descriptions for HAR

no code implementations22 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.

Human Activity Recognition Quantization

ContextMix: A context-aware data augmentation method for industrial visual inspection systems

1 code implementation18 Jan 2024 Hyungmin Kim, Donghun Kim, Pyunghwan Ahn, Sungho Suh, Hansang Cho, Junmo Kim

With the minimal additional computation cost of image resizing, ContextMix enhances performance compared to existing augmentation techniques.

Data Augmentation Object Recognition

CoSS: Co-optimizing Sensor and Sampling Rate for Data-Efficient AI in Human Activity Recognition

no code implementations3 Jan 2024 Mengxi Liu, Zimin Zhao, Daniel Geißler, Bo Zhou, Sungho Suh, Paul Lukowicz

Recent advancements in Artificial Neural Networks have significantly improved human activity recognition using multiple time-series sensors.

Human Activity Recognition Time Series

The Power of Training: How Different Neural Network Setups Influence the Energy Demand

no code implementations3 Jan 2024 Daniel Geißler, Bo Zhou, Mengxi Liu, Sungho Suh, Paul Lukowicz

This work offers a heuristic evaluation of the effects of variations in machine learning training regimes and learning paradigms on the energy consumption of computing, especially HPC hardware with a life-cycle aware perspective.

Transfer Learning

Remaining Useful Life Prediction of Lithium-ion Batteries using Spatio-temporal Multimodal Attention Networks

no code implementations29 Oct 2023 Sungho Suh, Dhruv Aditya Mittal, Hymalai Bello, Bo Zhou, Mayank Shekhar Jha, Paul Lukowicz

The proposed ST-MAN is to capture the complex spatio-temporal dependencies in the battery data, including the features that are often neglected in existing works.

In situ Fault Diagnosis of Indium Tin Oxide Electrodes by Processing S-Parameter Patterns

no code implementations16 Aug 2023 Tae Yeob Kang, Haebom Lee, Sungho Suh

In the field of optoelectronics, indium tin oxide (ITO) electrodes play a crucial role in various applications, such as displays, sensors, and solar cells.

Dimensionality Reduction Fault Detection

Worker Activity Recognition in Manufacturing Line Using Near-body Electric Field

no code implementations7 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.

Activity Recognition Time Series

Two-stage Early Prediction Framework of Remaining Useful Life for Lithium-ion Batteries

no code implementations7 Aug 2023 Dhruv Mittal, Hymalai Bello, Bo Zhou, Mayank Shekhar Jha, Sungho Suh, Paul Lukowicz

Early prediction of remaining useful life (RUL) is crucial for effective battery management across various industries, ranging from household appliances to large-scale applications.

Management

Selecting the motion ground truth for loose-fitting wearables: benchmarking optical MoCap methods

1 code implementation21 Jul 2023 Lala Shakti Swarup Ray, Bo Zhou, Sungho Suh, Paul Lukowicz

To help smart wearable researchers choose the optimal ground truth methods for motion capturing (MoCap) for all types of loose garments, we present a benchmark, DrapeMoCapBench (DMCB), specifically designed to evaluate the performance of optical marker-based and marker-less MoCap.

Benchmarking

Proxy Anchor-based Unsupervised Learning for Continuous Generalized Category Discovery

1 code implementation ICCV 2023 Hyungmin Kim, Sungho Suh, Daehwan Kim, Daun Jeong, Hansang Cho, Junmo Kim

Existing methods for novel category discovery are limited by their reliance on labeled datasets and prior knowledge about the number of novel categories and the proportion of novel samples in the batch.

Class Incremental Learning Incremental Learning +1

ClothFit: Cloth-Human-Attribute Guided Virtual Try-On Network Using 3D Simulated Dataset

no code implementations24 Jun 2023 Yunmin Cho, Lala Shakti Swarup Ray, Kundan Sai Prabhu Thota, Sungho Suh, Paul Lukowicz

The proposed method utilizes a U-Net-based network architecture that incorporates cloth and human attributes to guide the realistic virtual try-on synthesis.

Attribute Decoder +1

MeciFace: Mechanomyography and Inertial Fusion-based Glasses for Edge Real-Time Recognition of Facial and Eating Activities

no code implementations19 Jun 2023 Hymalai Bello, Sungho Suh, Bo Zhou, Paul Lukowicz

The increasing prevalence of stress-related eating behaviors and their impact on overall health highlights the importance of effective and ubiquitous monitoring systems.

Facial Expression Recognition Management +1

CaptAinGlove: Capacitive and Inertial Fusion-Based Glove for Real-Time on Edge Hand Gesture Recognition for Drone Control

no code implementations7 Jun 2023 Hymalai Bello, Sungho Suh, Daniel Geißler, Lala Ray, Bo Zhou, Paul Lukowicz

We present CaptAinGlove, a textile-based, low-power (1. 15Watts), privacy-conscious, real-time on-the-edge (RTE) glove-based solution with a tiny memory footprint (2MB), designed to recognize hand gestures used for drone control.

Hand Gesture Recognition Hand-Gesture Recognition +1

Chemical Property-Guided Neural Networks for Naphtha Composition Prediction

no code implementations2 Jun 2023 Chonghyo Joo, Jeongdong Kim, Hyungtae Cho, Jaewon Lee, Sungho Suh, Junghwan Kim

In this paper, we propose a neural network framework that utilizes chemical property information to improve the performance of naphtha composition prediction.

Unsupervised Statistical Feature-Guided Diffusion Model for Sensor-based Human Activity Recognition

no code implementations30 May 2023 Si Zuo, Vitor Fortes Rey, Sungho Suh, Stephan Sigg, Paul Lukowicz

A key problem holding up progress in wearable sensor-based HAR, compared to other ML areas, such as computer vision, is the unavailability of diverse and labeled training data.

Generative Adversarial Network Human Activity Recognition +1

FieldHAR: A Fully Integrated End-to-end RTL Framework for Human Activity Recognition with Neural Networks from Heterogeneous Sensors

no code implementations22 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.

Human Activity Recognition

Learning Graph Patterns of Reflection Coefficient for Non-destructive Diagnosis of Cu Interconnects

no code implementations20 Apr 2023 Tae Yeob Kang, Haebom Lee, Sungho Suh

Fault detection and diagnosis of the interconnects are crucial for prognostics and health management (PHM) of electronics.

Ensemble Learning Fault Detection +2

A Knowledge Distillation framework for Multi-Organ Segmentation of Medaka Fish in Tomographic Image

no code implementations24 Feb 2023 Jwalin Bhatt, Yaroslav Zharov, Sungho Suh, Tilo Baumbach, Vincent Heuveline, Paul Lukowicz

Morphological atlases are an important tool in organismal studies, and modern high-throughput Computed Tomography (CT) facilities can produce hundreds of full-body high-resolution volumetric images of organisms.

Computed Tomography (CT) Image Segmentation +4

AI-KD: Adversarial learning and Implicit regularization for self-Knowledge Distillation

no code implementations20 Nov 2022 Hyungmin Kim, Sungho Suh, SungHyun Baek, Daehwan Kim, Daun Jeong, Hansang Cho, Junmo Kim

Our model not only distills the deterministic and progressive knowledge which are from the pre-trained and previous epoch predictive probabilities but also transfers the knowledge of the deterministic predictive distributions using adversarial learning.

Self-Knowledge Distillation

Learning from the Best: Contrastive Representations Learning Across Sensor Locations for Wearable Activity Recognition

no code implementations4 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.

Activity Recognition Wearable Activity Recognition

TASKED: Transformer-based Adversarial learning for human activity recognition using wearable sensors via Self-KnowledgE Distillation

no code implementations14 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.

Human Activity Recognition Self-Knowledge Distillation

Estimation of 3D Body Shape and Clothing Measurements from Frontal- and Side-view Images

no code implementations28 May 2022 Kundan Sai Prabhu Thota, Sungho Suh, Bo Zhou, Paul Lukowicz

The estimation of 3D human body shape and clothing measurements is crucial for virtual try-on and size recommendation problems in the fashion industry but has always been a challenging problem due to several conditions, such as lack of publicly available realistic datasets, ambiguity in multiple camera resolutions, and the undefinable human shape space.

Virtual Try-on

Adversarial Deep Feature Extraction Network for User Independent Human Activity Recognition

no code implementations23 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.

Decoder Human Activity Recognition

Generalized multiscale feature extraction for remaining useful life prediction of bearings with generative adversarial networks

1 code implementation26 Sep 2021 Sungho Suh, Paul Lukowicz, Yong Oh Lee

The experimental results show that the proposed feature extraction method can effectively predict the RUL and outperforms the conventional RUL prediction approaches based on deep neural networks.

Sequential Targeting: an incremental learning approach for data imbalance in text classification

no code implementations20 Nov 2020 Joel Jang, Yoonjeon Kim, Kyoungho Choi, Sungho Suh

Classification tasks require a balanced distribution of data to ensure the learner to be trained to generalize over all classes.

General Classification Incremental Learning +2

Discriminative feature generation for classification of imbalanced data

1 code implementation24 Oct 2020 Sungho Suh, Paul Lukowicz, Yong Oh Lee

In this paper, we propose a novel supervised discriminative feature generation (DFG) method for a minority class dataset.

Classification Data Augmentation +2

Two-stage generative adversarial networks for document image binarization with color noise and background removal

1 code implementation20 Oct 2020 Sungho Suh, Jihun Kim, Paul Lukowicz, Yong Oh Lee

Document image enhancement and binarization methods are often used to improve the accuracy and efficiency of document image analysis tasks such as text recognition.

Binarization Decoder +1

Fusion of Global-Local Features for Image Quality Inspection of Shipping Label

no code implementations26 Aug 2020 Sungho Suh, Paul Lukowicz, Yong Oh Lee

These results are expected to improve the shipping address recognition and verification system by applying different image preprocessing steps based on the classified conditions.

object-detection Object Detection

Cannot find the paper you are looking for? You can Submit a new open access paper.