I am Tao Zhang (张韬), currently a PhD student at Wuhan University, under the guidance of Prof. Shunping Ji. Additionally, I am currently interning at Seed, ByteDance, where I am advised by Dr. Xiangtai Li and Dr. Jiashi Feng. Prior to this, I completed internships at the Y-Tech Lab of Kuaishou Technology (advised by Xingye Tian) and Skywork AI (advised by Dr. Xiangtai Li and Prof. ShuiCheng YAN). I obtained my master’s degree from Wuhan University under the supervision of Prof. Shunping Ji.

My research interests encompass:

  1. Image and video understanding tasks.
  2. Multi-modal learning with Large Language Models (LLMs).
  3. Remote Sensing.

🔥 News

  • 2024.12:  🎉🎉 PCM is accepted by AAAI 2025.
  • 2024.06:  🎉🎉 OMG-LLaVA is accepted by NeurIPS 2024.
  • 2024.04:  🎉🎉 DVIS-DAQ is accepted by ECCV 2024.
  • 2023.08:  🎉🎉 We achieve 1st place in the VIS Track of the 5th LSVOS challenge at ICCV 2023.
  • 2023.07:  🎉🎉 DVIS is accepted by ICCV 2023.
  • 2023.05:  🎉🎉 We achieve 1st place in the VPS Track of the PVUW challenge at CVPR 2023.

📝 Publications

Full Publications can be found in Google Scholar.

Code can be found in github.

Main Publications:

  • OMG-LLaVA: Bridging Image-level, Object-level, Pixel-level Reasoning and Understanding, Tao Zhang, Xiangtai Li, Hao Fei, Haobo Yuan, Shengqiong Wu, Shunping Ji, Chen Change Loy, Shuicheng Yan, NeurIPS 2024 | Code
  • Point Could Mamba: Point Cloud Learning via State Space Model, Tao Zhang, Haobo Yuan, Lu Qi, Jiangning Zhang, Qianyu Zhou, Shunping Ji, Shuicheng Yan, Xiangtai Li, AAAI 2025 | Code
  • Improving Video Segmentation via Dynamic Anchor Queries, Yikang Zhou*, Tao Zhang*, Shunping Ji, Shuicheng Yan, Xiangtai Li, ECCV 2024 | Code
  • DVIS++: Improved Decoupled Framework for Universal Video Segmentation, Tao Zhang, Xingye Tian, Yikang Zhou, Shunping Ji, Xuebo Wang, Yuan Zhang, Pengfei Wan, Zhongyuan Wang, Yu Wu, arxiv | Code
  • DVIS: Decoupled Video Instance Segmentation Framework, Tao Zhang, Xingye Tian, Yu Wu, Shunping Ji, Xuebo Wang, Yuan Zhang, Pengfei Wan, ICCV 2023 | Code
  • E2EC: An End-to-End Contour-based Method for High-Quality High-Speed Instance Segmentation, Tao Zhang, Shiqing Wei, Shunping Ji, CVPR 2022 | Code
  • 🎖 Honors and Awards

    • 2023.08 Winner of the VIS Track of the 5th LSVOS challenge at ICCV 2023.
    • 2023.05 Winner of the VPS Track of the PVUW challenge at CVPR 2023.
    • Conference Reviewer for ICML 2025, CVPR 2025, ICLR 2025, AAAI 2025, NeurIPS 2024, CVPR 2024, ECCV 2024, CVPR 2023, ICCV 2023 and Journal Reviewer For IEEE-TCSVT and IEEE-TGRS.

    📖 Educations

    • 2023.09 - now, PhD in Wuhan University.
    • 2020.09 - 2023.06, Master in Wuhan University.
    • 2016.09 - 2020.06, Bachelor in Northeastern University.

    💻 Internships

    • 2022.01 - 2024.02, Y-Tech Lab of Kuaishou technology, China.
    • 2024.02 - 2024.06, Skywork AI, Singapore.
    • 2024.09 - now, Seed, ByteDance, China.