정보통신대학 - 정보통신대학

  • 조교수 인공지능 HW 및 시스템
  • 고종환 홈페이지 바로가기
    Lab 지능 저전력 음성/영상 시스템 연구실
CV1

관심분야

온-디바이스 머신러닝, AI 프로세서, 뉴로모픽 컴퓨팅

학력

  • 2000-2004: 서울대학교 컴퓨터공학, 기계항공공학 학사 (복수전공),
  • 2004-2006: 서울대학교 전기컴퓨터공학 석사,
  • 2013-2018: Georgia Institute of Technology 전자컴퓨터공학 박사

약력/경력

  • 2006-2019: 국방과학연구소 선임연구원,
  • 2017: Microsoft Research (Research intern),
  • 2019 ~: 성균관대 전자전기공학부 조교수

학술지 논문

  • (2023)  Highly Reliable 3D Channel Memory and Its Application in a Neuromorphic Sensory System for Hand Gesture Recognition.  ACS NANO.  17,  24
  • (2023)  Adaptive Channel Scheduling for Acceleration and Fine Control of RNN-Based Image Compression.  IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES.  E106.A,  9
  • (2023)  Data-Driven Design of Electrically Conductive Nanocomposite Materials: A Case Study of Acrylonitrile-Butadiene-Styrene/Carbon Nanotube Binary Composites.  ADVANCED INTELLIGENT SYSTEMS.  5,  2
  • (2023)  Room-Temperature-Processable Highly Reliable Resistive Switching Memory with Reconfigurability for Neuromorphic Computing and Ultrasonic Tissue Classification.  ADVANCED FUNCTIONAL MATERIALS.  1,  1
  • (2023)  An overhead-free region-based JPEG framework for task-driven image compression.  PATTERN RECOGNITION LETTERS.  165,  1
  • (2022)  A Reconfigurable Neural Architecture for EdgeCloud Collaborative Real-Time Object Detection.  IEEE INTERNET OF THINGS JOURNAL.  9,  23
  • (2022)  Micro-ultrasonic Assessment of Early Stage Clot Formation and Whole Blood Coagulation Using an All-Optical Ultrasound Transducer and Adaptive Signal Processing Algorithm.  ACS SENSORS.  7,  10
  • (2022)  Epidermal piezoresistive structure with deep learning-assisted data translation.  NPJ FLEXIBLE ELECTRONICS.  6,  1
  • (2022)  Scalable Color Quantization for Task-Centric Image Compression.  ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS.  1,  1
  • (2022)  Multi-Prediction Compression: An Efficient and Scalable Memory Compression Framework for GP-GPU.  IEEE COMPUTER ARCHITECTURE LETTERS.  21,  2
  • (2022)  VWC-SDK: Convolutional Weight Mapping Using Shifted and Duplicated Kernel With Variable Windows and Channels.  IEEE JOURNAL ON EMERGING AND SELECTED TOPICS IN CIRCUITS AND SYSTEMS.  12,  2
  • (2022)  Adaptive weight-bit inversion for state error reduction for robust and efficient deep neural network inference using MLC NAND Flash.  JOURNAL OF SYSTEMS ARCHITECTURE.  124,  1
  • (2022)  ORVAE: One-Class Residual Variational Autoencoder for Voice Activity Detection in Noisy Environment.  NEURAL PROCESSING LETTERS.  54,  3
  • (2021)  Binary-Classifiers-Enabled Filters for Semi-Supervised Learning.  IEEE ACCESS.  9,  -
  • (2021)  Target-Dependent Scalable Image Compression Using a Reconfigurable Recurrent Neural Network.  IEEE ACCESS.  9, 
  • (2021)  A Charge-Domain Scalable-Weight In-Memory Computing Macro With Dual-SRAM Architecture for Precision-Scalable DNN Accelerators.  IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS.  68,  8
  • (2021)  Robust detection of small and dense objects in images from autonomous aerial vehicles.  ELECTRONICS LETTERS.  1,  1
  • (2021)  Segmentation of Points in the Future: Joint Segmentation and Prediction of a Point Cloud.  IEEE ACCESS.  9, 
  • (2019)  CAMEL: An Adaptive Camera With Embedded Machine Learning-Based Sensor Parameter Control.  IEEE JOURNAL ON EMERGING AND SELECTED TOPICS IN CIRCUITS AND SYSTEMS.  9,  3
  • (2018)  3-D Stacked Image Sensor With Deep Neural Network Computation.  IEEE SENSORS JOURNAL.  18,  10

특허/프로그램

  • 경량화된 피험자 적응 기법.  C-2023-046625.  20231023.  대한민국
  • Integrated high-speed image sensor and operation method thereof.  17/500538.  20230822.  미국
  • 인공지능 신경망을 이용한 음원 분류 및 위치 추적 방법 및 시스템(The mothod and system for sound classification and location tracking using artificial intelligence nural network).  10-2556-8100000.  20230713.  대한민국
  • 심층신경망을 이용한 음성 신호의 발생 방향 탐지 방법.  10-2020-0032737.  20211116.  대한민국
  • 다중 통신 제어 장치 및 방법.  10-2019-0017121.  20200925.  대한민국
  • 이중 연결 관리 장치 및 방법(APPARATUS AND METHOD FOR MANAGING DUAL CONNECTIVITY).  10-2019-0016926.  20200508.  대한민국
  • 통신 장치 제어 시스템 및 그 제어 방법.  10-2019-0007021.  20200325.  대한민국
  • 전자파 인체영향 분석을 위한 수치 팬텀 모델 제작 방법(Numerical phantom model manufacturing method for analyzing on human effects of Radio Frequency).  10-1486-8470000.  20150129.  대한민국

학술회의논문

  • (2023)  Mip-Grid: Anti-aliased Grid Representations for Neural Radiance Fields.  Conference on Neural Information Processing Systems.  미국
  • (2023)  DCR: Decomposition-Aware Column Re-Mapping for Stuck-At-Fault Tolerance in ReRAM Arrays.  IEEE International Conference on Computer Design.  미국
  • (2023)  FFNeRV: Flow-Guided Frame-Wise Neural Representations for Videos.  ACM Multimedia Conference.  캐나다
  • (2023)  Kernel Shape Control for Row-Efficient Convolution on Processing-In-Memory Arrays.  IEEE/ACM International Conference on Computer-Aided Design.  미국
  • (2023)  HandR2N2: Iterative 3D Hand Pose Estimation Using a Residual Recurrent Neural Network.  International Conference on Computer Vision.  프랑스
  • (2023)  Multi-Scale Bidirectional Recurrent Network with Hybrid Correlation for Point Cloud Based Scene Flow Estimation.  International Conference on Computer Vision.  프랑스
  • (2023)  PAIRS: Pruning-AIded Row-Skipping for SDK-Based Convolutional Weight Mapping in Processing-In-Memory Architectures.  IEEE International Symposium on Low-Power Electronics and Design.  오스트리아
  • (2023)  Weight-Aware Activation Mapping for Energy-Efficient Convolution on PIM Arrays.  IEEE International Symposium on Low-Power Electronics and Design.  오스트리아
  • (2023)  Masked Wavelet Representation for Compact Neural Radiance Fields.  Conference on Computer Vision and Pattern Recognition.  미국
  • (2022)  Neural Residual Flow Fields for Efficient Video Representations.  Asian Conference on Computer Vision.  마카오
  • (2022)  Element-wise Partial Product Quantization for Efficient Deep Learning Accelerators.  "International Conference On Consumer Electronics Asia(ICCE Asia) Asia".  대한민국
  • (2022)  Bi-PointFlowNet: Bidirectional Learning for Point Cloud Based Scene Flow Estimation.  European Conference on Computer Vision.  이스라엘
  • (2022)  Streamable Neural Fields.  European Conference on Computer Vision.  이스라엘
  • (2022)  An Efficient Systolic Array with Variable Data Precision and Dimension Support.  International SoC Design Conference (ISOCC).  대한민국
  • (2022)  NAS-VAD: Neural Architecture Search for Voice Activity Detection.  Conference of the International Speech Communication Association.  대한민국
  • (2022)  ADA-VAD: Unpaired Adversarial Domain Adaptation for Noise-Robust Voice Activity Detection.  IEEE International Conference on Acoustics, Speech and Signal Processing.  싱가포르
  • (2022)  VW-SDK: Efficient Convolutional Weight Mapping Using Variable Windows for Processing-In-Memory Architectures.  Design Automation and Test in Europe Conference.  벨기에
  • (2021)  A Splittable DNN-Based Object Detector for Edge-Cloud Collaborative Real-Time Video Inference.  IEEE International Conference on Advanced Video and Signal-Based Surveillance.  미국
  • (2021)  HandFoldingNet: A 3D Hand Pose Estimation Network Using Multiscale-Feature Guided Folding of a 2D Hand Skeleton.  International Conference on Computer Vision.  미국
  • (2021)  A Sub-Milliwatt and Sub-Millisecond 3-D Gaze Estimator for Ultra Low-Power AR Applications.  Ubicomp/ISWC-Eyeware '21.  미국