TKTaemin Kim

Taemin Kim

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Taemin Kim has a total of 92 co-authors. Here are the top co-authors by number of shared publications:

  • Ki Jun Yu

    School of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of Korea

    5 shared publications
  • Kyubeen Kim

    Functional Bio-integrated Electronics and Energy Management Lab, School of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul, Republic of Korea

    3 shared publications
  • Mingyu Sang

    Department of Electronic Engineering Gachon University, 1342, Seongnam-daero Sujeong-gu, Seongnam-si Gyeonggi-do, Republic of Korea

    3 shared publications
  • Byung Gwan Son

    Digital Signal Processing & Artificial Intelligence Lab, School of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of Korea

    2 shared publications
  • Dosik Hwang

    Yonsei University

    2 shared publications
  • Geonhui Son

    Medical Artificial Intelligence Lab, School of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of Korea

    2 shared publications

Latest Publications

  • Open Access

    An AI-driven, wearable, conformal ring system for real-time and user-independent sign language interpretation

    by Jaejin Park, Yejee Shin, In Sik Min, Yeeun Lee, Jiwon Lee, Jung-Hoon Hong, Sunho Park, Sang Hoon Park, Junhyeong Baek, Taemin Kim, et al.

    Sign language translation systems have long aimed to bridge the communication between signers and nonsigners. However, preliminary systems rely on glove-type wearables or wired sensor arrays, which constrain hand movement, reduce comfort, and require nonpersonalized sensor positions that limit adaptability across users. Here, we introduce a wirelessly connected, ring-type sign language translator (WRSLT) designed to overcome these limitations by enabling full finger mobility through independent sensor rings and multilink communication. The system supports static and dynamic gesture detection using selected fingers via quantitative relevance analysis and achieves robust user-independent performance without per-user calibration. WRSLT demonstrated high recognition accuracy on large-scale datasets comprising 100 American Sign Language and 100 International Sign Language words, achieving 88.3 and 88.5% accuracy, respectively, under unseen-user conditions (i.e., test users not included in model training). Furthermore, a custom sequential word detection framework enables sentence-level translation from continuous signing input without requiring separate training on entire sentence structures.


  • Top Cited Papers
    Open Access

    Transparent, metal-free PEDOT:PSS neural interfaces for simultaneous recording of low-noise electrophysiology and artifact-free two-photon imaging

    by Hyun Woo Kim, Jiwon Kim, Jong Youl Kim, Kyubeen Kim, Ju Young Lee, Taemin Kim, Shinil Cho, Jong Bin An, Hyun Jae Kim, Lulu Sun, et al.

    Simultaneous two-photon imaging and electrophysiological recordings offer considerable potential for advancing neurological research and therapies. However, traditional metal-based neural interfaces suffer from photoelectric artifacts, while existing transparent implants rely on opaque interconnect lines to address conductivity limitations. Herein, we developed an optically transparent poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) neural electrode array with transparent electrodes and interconnect lines. Through a formamide, phosphoric acid, and ethylene glycol treatment, the metal-free PEDOT:PSS array achieved an impedance of 45.8 kΩ (at 1 kHz) even with a 20 × 20 µm² size. This advanced performance surpasses previous metal-free transparent neural interfaces and facilitates precise electrophysiological recordings, including extracellular action potentials and low-noise local field potentials. In vivo experiments demonstrated artifact-free two-photon imaging and reliable neural signal acquisition, while biocompatibility tests confirmed negligible cytotoxicity or immune responses. The developed metal-free PEDOT:PSS array provides a robust platform for neural recording and bioimaging, representing an advancement in transparent neural interface technology and integrated optical modalities.


  • Top Cited Papers

    Vertical full-colour micro-LEDs via 2D materials-based layer transfer

    by Jiho Shin, Hyunseok Kim, Suresh Sundaram, Junseok Jeong, Bo-In Park, Celesta S. Chang, Joonghoon Choi, Taemin Kim, Mayuran Saravanapavanantham, Kuangye Lu, et al.

    Micro-LEDs (µLEDs) have been explored for augmented and virtual reality display applications that require extremely high pixels per inch and luminance1,2. However, conventional manufacturing processes based on the lateral assembly of red, green and blue (RGB) µLEDs have limitations in enhancing pixel density3,4,5,6. Recent demonstrations of vertical µLED displays have attempted to address this issue by stacking freestanding RGB LED membranes and fabricating top-down7,8,9,10,11,12,13,14, but minimization of the lateral dimensions of stacked µLEDs has been difficult. Here we report full-colour, vertically stacked µLEDs that achieve, to our knowledge, the highest array density (5,100 pixels per inch) and the smallest size (4 µm) reported to date. This is enabled by a two-dimensional materials-based layer transfer technique15,16,17,18 that allows the growth of RGB LEDs of near-submicron thickness on two-dimensional material-coated substrates via remote or van der Waals epitaxy, mechanical release and stacking of LEDs, followed by top-down fabrication. The smallest-ever stack height of around 9 µm is the key enabler for record high µLED array density. We also demonstrate vertical integration of blue µLEDs with silicon membrane transistors for active matrix operation. These results establish routes to creating full-colour µLED displays for augmented and virtual reality, while also offering a generalizable platform for broader classes of three-dimensional integrated devices.


  • Top Cited Papers
    Open Access

    Ultrathin crystalline-silicon-based strain gauges with deep learning algorithms for silent speech interfaces

    by Taemin Kim, Yejee Shin, Kyowon Kang, Kiho Kim, Gwanho Kim, Yunsu Byeon, Hwayeon Kim, Yuyan Gao, Jeong Ryong Lee, Geonhui Son, et al.

    A wearable silent speech interface (SSI) is a promising platform that enables verbal communication without vocalization. The most widely studied methodology for SSI focuses on surface electromyography (sEMG). However, sEMG suffers from low scalability because of signal quality-related issues, including signal-to-noise ratio and interelectrode interference. Hence, here, we present a novel SSI by utilizing crystalline-silicon-based strain sensors combined with a 3D convolutional deep learning algorithm. Two perpendicularly placed strain gauges with minimized cell dimension (2) could effectively capture the biaxial strain information with high reliability. We attached four strain sensors near the subject’s mouths and collected strain data of unprecedently large wordsets (100 words), which our SSI can classify at a high accuracy rate (87.53%). Several analysis methods were demonstrated to verify the system’s reliability, as well as the performance comparison with another SSI using sEMG electrodes with the same dimension, which exhibited a relatively low accuracy rate (42.60%).


  • Open Access

    Ultra-thin crystalline silicon-based strain gauges with deep learning algorithms for silent speech interfaces

    by Ki Jun Yu, Taemin Kim, Yejee Shin, Kyowon Kang, Kiho Kim, Gwanho Kim, Yunsu Byeon, Hwayeon Kim, Yuyan Gao, Jihyun Kim, et al.

    A wearable silent speech interface (SSI) is a promising platform that enables verbal communication without vocalization. The most widely studied methodology for SSI focuses on surface electromyography (sEMG). However, sEMG suffers from low scalability because of signal quality-related issues, including signal-to-noise ratio and interelectrode interference. Hence, in this study, we present a novel SSI by utilizing crystalline-silicon-based strain sensors combined with a 3D convolutional deep learning algorithm. Two perpendicularly placed strain gauges with minimized cell dimension (< 0.1 mm2) could capture the biaxial strain information with high reliability. We attached four strain sensors near the subject’s mouths and collected strain data of unprecedently large wordsets (100 words), which our SSI can classify at a high accuracy rate (87.53%). Several analysis methods were demonstrated to verify the system’s reliability, as well as the performance comparison with another SSI using sEMG electrodes with the same dimension, which exhibited an accuracy rate of 35.00%.