Jiayi Ni   |   倪嘉怡

Enrolled at 16, I am a sophomore at Peking University, advised by Prof. Yitao Liang, Prof. Hang Zhao and Prof. Shanghang Zhang.

My overarching objective is to build an open-world generalist agent with complex perception, planning and control capabilities.

Email  /  Google Scholar  /  Github /  Twitter /  Youtube /  WeChat

profile photo
Publications


Distribution-Aware Continual Test Time Adaptation for Semantic Segmentation
Jiayi Ni*, Senqiao Yang*, Jiaming Liu†, Xiaoqi Li, Wenyu Jiao, Ran Xu, Zehui Chen, Yi Liu, Shanghang Zhang
Paper / Video / Code
ICRA 2024

Robust Continual Test Time Adaptation via Visual Domain Adapter
PKU_BLV_LAB Team, with Ran Xu, Peidong Jia, Senqiao Yang, Jiaming Liu, Zehui Chen, Feng Zhao, Yandong Guo, Shanghang Zhang
Technical Report / Award / Video / Talk / Workshop / Challenge Page
1st Place in ICCV 2023 Challenge Track


Recent Talks

  • [Oct 15] "Distribution-Aware Continual Test Time Adaptation for Semantic Segmentation" at PKU-IAI Annual Symposium and Tech Day [Poster] (Rank: 1st in Creativity; 2nd in Audience approval)
  • [Oct 2] Winner presentation on ICCV 2023 Visual Continual Learning Workshop [Link/Video]


  • Services

  • Reviewer: IEEE Transactions on Neural Networks and Learning Systems (TNNLS 2024).
  • Tong Class leader of Yuanpei Information & Data & Intelligence Society; hosted dozens of experience-sharing sessions on research and course study.

  • Selected Awards

  • 2023: SenseTime Scholarship Most Promising Award (Youngest, 18 yrs)
  • 2023: Peking University Research Excellence Award (<0.5%)
  • 2023: Yicong Huang Scholarship (¥8000 RMB)
  • 2023: Jun Yuan Scholarship (¥7000 RMB)
  • 2023: 1st Place in ICCV 2023 Challenge Track ($1000 USD) 🏅
  • 2022: Peking University Student Excellence Award
  • 2022: Peking University Scholarship (¥4000 RMB)
  • 2021: Peking University Freshman Scholarship (¥10000 RMB)
  • 2019: 1st Prize in Real World Design Challenge (China)

  • This template is a modification to Yiran Geng and Jon Barron's website.