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

被引:2
|
作者
Ray, Lala Shakti Swarup [1 ]
Zhou, Bo [1 ,2 ]
Suh, Sungho [1 ,2 ]
Krupp, Lars [1 ,2 ]
Rey, Vitor Fortes [1 ,2 ]
Lukowicz, Paul [1 ,2 ]
机构
[1] DFKI, Kaiserslautern, Germany
[2] RPTU Kaiserslautern Landau, Kaiserslautern, Germany
关键词
HAR; Generative model; GPT; Pressure Sensor; Synthetic dataset;
D O I
10.1109/PerComWorkshops59983.2024.10503379
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In human activity recognition (HAR), the availability of substantial ground truth is necessary for training efficient models. However, acquiring ground pressure data through physical sensors itself can be cost-prohibitive, time-consuming. To address this critical need, we introduce Text-to-Pressure (T2P), a framework designed to generate extensive ground pressure sequences from textual descriptions of human activities using deep learning techniques. 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. We achieved comparable performance on the consistency between text and generated motion with an R squared value of 0.722, Masked R squared value of 0.892, and FID score of 1.83. Additionally, we trained a HAR model with the the synthesized data and evaluated it on pressure dynamics collected by a real pressure sensor which is on par with a model trained on only real data. Combining both real and synthesized training data increases the overall macro F1 score by 5.9 percent.
引用
收藏
页码:461 / 464
页数:4
相关论文
共 50 条
  • [21] GIFT: a tool for generating free text reports from encoded data
    Panzarasa, Silvia
    Quaglini, Silvana
    Pessina, Mauro
    Cavallini, Anna
    Micieli, Giuseppe
    11TH MEDITERRANEAN CONFERENCE ON MEDICAL AND BIOLOGICAL ENGINEERING AND COMPUTING 2007, VOLS 1 AND 2, 2007, 16 (1-2): : 152 - +
  • [22] Text2Simulate: A Scientific Knowledge Visualization Technique for Generating Visual Simulations from Textual Knowledge
    Ige, Ifeoluwatayo A.
    Oladejo, Bolanle F.
    INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS, 2023, 14 (02) : 14 - 22
  • [23] ARTIFICIAL INTELLIGENCE AND CREATIVITY FOR GENERATING ARCHITECTURAL IMAGES FROM TEXTUAL DESCRIPTIONS IN MIDJOURNEY. EMULATING LOUIS I. KAHN
    Molina-Siles, Pedro
    Ribera, Manuel Gimenez
    EGA-REVISTA DE EXPRESION GRAFICA ARQUITECTONICA, 2023, 28 (49): : 238 - 251
  • [24] The Informational Value of Segment Data Disaggregated by Underlying Industry: Evidence from the Textual Features of Business Descriptions
    Song, Shiwon
    ACCOUNTING REVIEW, 2021, 96 (06): : 361 - 396
  • [25] Toward Explainable Recommendations: Generating Review Text from Multicriteria Evaluation Data
    Suzuki, Takafumi
    Oyama, Satoshi
    Kurihara, Masahito
    2018 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2018, : 3549 - 3551
  • [26] New Insights from Existing Sequence Data: Generating Breakthroughs without a Pipette
    Plocik, Alex M.
    Graveley, Brenton R.
    MOLECULAR CELL, 2013, 49 (04) : 605 - 617
  • [27] VTM-GAN: video-text matcher based generative adversarial network for generating videos from textual description
    Mehmood R.
    Bashir R.
    Giri K.J.
    International Journal of Information Technology, 2024, 16 (1) : 221 - 236
  • [28] Ground Deformation Data from GEER Investigations of Ridgecrest Earthquake Sequence
    Brandenberg, Scott J.
    Stewart, Jonathan P.
    Wang, Pengfei
    Nweke, Chukwuebuka C.
    Hudson, Kenneth
    Goulet, Christine A.
    Meng, Xiaofeng
    Davis, Craig A.
    Ahdi, Sean K.
    Hudson, Martin B.
    Donnellan, Andrea
    Lyzenga, Gregory
    Pierce, Marlon
    Wang, Jun
    Winters, Maria A.
    Delisle, Marie -Pierre
    Lucey, Joseph
    Kim, Yeulwoo
    Gallien, Timu W.
    Lyda, Andrew
    Yeung, J. Sean
    Issa, Omar
    Buckreis, Tristan
    Yi, Zhengxiang
    SEISMOLOGICAL RESEARCH LETTERS, 2020, 91 (04) : 2024 - 2034
  • [29] From Textual Data to Theoretical Insights: Introducing and Applying the Word-Text-Topic Extraction Approach
    Jung, Jaewoo
    Zhou, Wenjun
    Smith, Anne D.
    ORGANIZATIONAL RESEARCH METHODS, 2024,
  • [30] A novel textual data augmentation method for identifying comparative text from user-generated content
    Wei, Na
    Zhao, Songzheng
    Liu, Jing
    Wang, Shenghui
    ELECTRONIC COMMERCE RESEARCH AND APPLICATIONS, 2022, 53