Dingfeng Shi (石鼎丰)
Dingfeng Shi (石鼎丰)
Computer vision(CV), large language model (LLM), video understanding and detection.

About Me

I am currently a researcher at OPPO Personal AI Lab, specializing in LLM agent research and its practical applications. I earned my PhD from Beihang University, under the mentorship of Haogang Zhu and Jian Cheng. Before that, I completed my Bachelor’s degree at Xidian University in 2019. Additionally, I spent approximately three years at JD Explore Academy, collaborating with Dacheng Tao, Qiong Cao, and Yujie Zhong.

My research focuses on computer vision, video understanding, object detection, and large language models (LLMs).

[Google Scholar], [Github], [Email].

News

  • [2023.02.27] One paper is accepted at CVPR 2023
  • [2022.09.15] One paper is accepted at NeurIPS 2022
  • [2022.07.09] One paper is accepted at ECCV 2022
  • [2020.03.06] One paper is accepted at ICME 2020

Publications

TaskCraft: Automated Generation of Agentic Tasks
Dingfeng Shi*, Jingyi Cao*, Qianben Chen*, Weichen Sun, Weizhen Li, Hongxuan Lu, Fangchen Dong, Tianrui Qin, King Zhu, Minghao Yang, Jian Yang, Ge Zhang, Jiaheng Liu, Changwang Zhang, Jun Wang, Yuchen Eleanor Jiang, Wangchunshu Zhou
Arxiv, 2025.
[PDF] [CODE]
Temporal Action Localization with Enhanced Instant Discriminability
Dingfeng Shi, Qiong Cao, Yujie Zhong, Shan An, Jian Cheng, Haogang Zhu, Dacheng Tao
Arxiv, 2023.
[PDF]
TriDet: Temporal Action Detection with Relative Boundary Modeling
Dingfeng Shi, Yujie Zhong, Qiong Cao, Lin Ma, Jia Li, Dacheng Tao
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
[PDF] [CODE]
Benefits of Permutation-equivariance in Auction Mechanisms
Tian Qin, Fengxiang He, Dingfeng Shi, Wenbing Huang, Dacheng Tao
Neural Information Processing Systems (NeurIPS), 2022.
[PDF] [CODE]
ReAct: Temporal Action Detection with Relational Action Queries
Dingfeng Shi, Yujie Zhong, Qiong Cao, Jing Zhang, Lin Ma, Jia Li and Dacheng Tao
European Conference on Computer Vision (ECCV), 2022.
[PDF] [CODE]
Reconstructing Part-level 3D Models from a Single Image
Dingfeng Shi, Yifan Zhao, Jia Li
IEEE International Conference on Multimedia and Expo (ICME), 2020.
[PDF]

Academic Services

I serve as a reviewer for journals such as TPAMI, TIP, TMM, TCSVT and PR, as well as conferences like CVPR, ICCV, ECCV, NeurIPS, and AAAI.