( $^\ast$: equal contribution, $^\dag$: corresponding author )

Journal Paper

• Mask-RCNN-CHFNet: An Improved Deep Learning for 3D Reverse Modeling of Iron Tailings (SiO2) Real-time Melting Process
  Yuefang Sun, Xinghui Hao, Yi Shi, Zhaozhuang Guo, Aimin Yang$^\dag$
  Submitted to Alexandria Engineering Journal, Volume 129, October 2025, Pages 1238-1257.
  

• Audio Matters Too! Enhancing Markerless Motion Capture with Audio Signals for String Performance Capture
  Yitong Jin$^\ast$, Zhiping Qiu$^\ast$, Yi Shi, Shuangpeng Sun, Chongwu Wang, Donghao Pan, Jiachen Zhao, Zhenghao Liang, Yuan Wang, Xiaobing Li, Feng Yu, Tao Yu$^\dag$, Qionghai Dai$^\dag$
  Submitted to ACM Transactions on Graphics (TOG), Volume 43, Issue 4.
  

Conference Paper

• ELGAR: Expressive Cello Performance Motion Generation for Audio
  Zhiping Qiu, Yitong Jin, Yuan Wang, Yi Shi, Chongwu Wang, Chao Tan, Xiaobing Li, Feng Yu, Tao Yu$^\dag$, Qionghai Dai$^\dag$
  Submitted to ACM SIGGRAPH 2025.
  

Early Project

• Dynamic Supplementary Heat System for Blast Furnace Slag
  Yi Shi, Xinghui Hao
  

  
  This is an intelligent manufacturing project at the intersection of artificial intelligence and metallurgy. The goal is to detect high-temperature molten material in the melt pool (with $SiO_2$ used in experiments), and to estimate its volume from the area observed in 2D images. Based on this estimation, the system aims to monitor and calculate the real-time heat compensation required, ultimately contributing to energy conservation and emission reduction.
  
  The following figure displays the pipeline of the system.