About

I am a PhD student at HKUST, advised by Prof. Qifeng CHEN. I previously obtained my Bachelor’s degree with highest honor from Computer Science program at Hongyi Honor College, Wuhan University (WHU) in China.

Currently, I am working as a research intern at the International Digital Economy Academy (IDEA). Before joining HKUST, I was fortunate to be advised by Prof. Qi DOU at CUHK on domain adaptation/generalization, Prof. Junsong YUAN at UB SUNY on action recognition and Prof. Tomizuka at UC Berkeley on autonomous driving.

Research Interests

I am broadly interested in machine learning, computer vision, and robotics, aiming to tackle following three challenges: model robustness / generalization, efficiency, and the lack of reasoning ability in deep learning. My dream is to build smarter AI that can percept, reason, and decide in a human-like way as well as with human-level robustness with minimal supervision.

My recent interests include,

  • Video Understanding, Processing & Synthesis
  • 3D-Vision / NeRF / CLIP / DDPM
  • Multimodal Learning
  • Domain Adaptation & Generalization

Publications

  • Federated Domain Generalization for Image Recognition via Cross-Client Style Transfer Junming Chen*, Meirui Jiang*, Qi Dou, Qifeng Chen (* Joint First)
    WACV 2023
    Arxiv | Code | Project

  • Real-time Streaming Video Denoising with Bidirectional Buffers
    Chenyang Qi*, Junming Chen*, Xin Yang, Qifeng Chen (* Equal contribution)
    ACM Multimedia 2022
    Arxiv | Code | Project

  • Anchor-Guided Online Meta Adaptation for Fast One-Shot Instrument Segmentation from Robotic Surgical Videos
    Zixu Zhao, Yueming Jin, Junming Chen, Bo Lu, Chi-Fai Ng, Yun-Hui Liu, Qi Dou, Peng-Ann Heng
    Medical Image Analysis 2021

  • Dynamic Graph CNN for Event-based Gesture Recognition
    Junming Chen, Jingjing Meng, Xinchao Wang, Junsong Yuan
    ISCAS 2020

News

Teaching

TA for COMP4471 & ELEC4240 Deep Learning in Computer Vision in Fall 2022 at HKUST.
TA for COMP5214 & ELEC5680 Advanced Deep Learning Architectures in Spring 2022 at HKUST.