Mingqi Gao

Mingqi Gao (高明琦)

PhD candidate in Computer Vision

Biography

Hi! I am Mingqi, a PhD candidate at University of Warwick and Southern University of Science and Technology (SUSTech). I am now doing research in the WMG Data Science Group and SUSTech VIP Lab, under the supervision of Prof. Jungong Han, Prof. Feng Zheng, and Prof. Giovanni Montana. Prior to this, I received my Master degree and Bachelor degree from Chongqing University and Inner Mongolia University respectively.

My research lies in Computer Vision and Deep Learning, particularly focusing on Video Object Segmentation (Awesome Works on ) and Vision-Language Models.

News

[06-2024] - Win the 1st place in the PVUW Challenge, MeViS Track, CVPR 2024 (report, talk, code)
[07-2023] - 1 paper on Referring Video Object Segmentation (paper, code) is accepted by ICCV 2023
[06-2023] - 2 papers on Referring (paper, code) and Visible-Thermal (paper, code) Video Object Segmentation are accepted by IEEE Transactions on Circuits, Systems for Video Technology
Loading...

SELECTED PUBLICATIONS

Please see my full publication list from my Google Scholar Profile

Guanghui Li*, Mingqi Gao* (Equal Contribution), Heng Liu, Xiantong Zhen, Feng Zheng. ''Learning cross-modal affinity for referring video object segmentation targeting limited samples''. IEEE/CVF International Conference on Computer Vision (ICCV), 2023. (code)
Jinyu Yang*, Mingqi Gao* (Equal Contribution), Runmin Cong, Chengjie Wang, Feng Zheng, Aleš Leonardis. ''Unveiling the Power of Visible-Thermal Video Object Segmentation''. IEEE Transactions on Circuits, Systems for Video Technology, 2023. (code)
Mingqi Gao, Jinyu Yang, Jungong Han, Ke Lu, Feng Zheng, Giovanni Montana. ''Decoupling Multimodal Transformers for Referring Video Object Segmentation''. IEEE Transactions on Circuits, Systems for Video Technology, 2023. (code)
Mingqi Gao, Feng Zheng, James Yu, Caifeng Shan, Guiguang Ding, and Jungong Han. ''Deep learning for video object segmentation: a review''. Artificial Intelligence Review, 2022. (project page, Awesome VOS works)
Mingqi Gao, Jungong Han, Feng Zheng, James Yu, and Giovanni Montana. ''Video object segmentation using point-based memory matching network''. Pattern Recognition, 2022.
Mingqi Gao, Hengxin Chen, Shenhai Zheng, and Bin Fang. ''Feature fusion and non-negative matrix factorization based active contours for texture segmentation''. Signal Processing, 2019.
Mingqi Gao, Hengxin Chen, Shenhai Zheng, and Bin Fang. ''A factorization based active contour model for texture segmentation''. IEEE International Conference on Image Processing (ICIP), 2016. (project page, code)

RESEARCH

My research lies in computer vision and deep learning. I have been working on the following research topics:

Active Contour / Level Set-based Image Segmentation

Iterative methodology for image segmentation, where region boundary is achieved by evolving a closed and smooth contour (red curve).

Typical ACMs and codes

My research work (FACM)

Deep Learning-based Video Object Segmentation (VOS)

Segment and track target objects on video frames, where targets could be specified via saliency/motion, manual annotations, or language.

Awesome VOS works

My survey paper

MISCELLANEOUS

Honors and Awards

1st Place in the Pixel-level Video Understanding in the Wild Challenge, MeViS Track, CVPR 2024 (report, talk, code).
National Scholarship for Graduate Students (top 2%), Ministry of Education, China, 2016.

Journal Reviews

IEEE Transactions on Circuits, Systems for Video Technology, Journal of Visual Communication and Image Representation, Pattern Recognition Letters, IET Image Processing

Conference Reviews

NeurIPS Track Datasets and Benchmarks