Jinlong Li

Jinlong Li (李金龙)

First-year PhD Student.
Multimedia and Human Understanding Group, MHUG

Email: jinlong.szu@gmail.com

Pushing a long-termist, strike the tough yet right things.

Biography

I am now a first-year PhD student at the Multimedia and Human Understanding Group (MHUG), under the supervision of Prof. Nicu Sebe. Previously, I worked at MeiTuan as a computer vision scientist, working closely with Dr. Lin Ma and Dr. Zequn Jie, focusing on 2D/3D Label-Efficient Detection and Segmentation. Before that, I obtained my B.Sc and M.Sc degree in Shenzhen University.

Research Interest

I work in the field of Computer Vision and Deep Learning. Recently, I focus on the following research topics:

Label-Efficient Learning on both 2D and 3D
Multi-Modal Perception and Reasoning Learning
Generative and Foundation Models

News

2024 Mar.: Served as a reviewer for ECCV 2023.
2024 Feb.: Served as a reviewer for ICML 2024.
2023 Aug.: Served as a reviewer for ICLR 2024.
2023 Mar.: Served as a reviewer for NeurIPS 2023.
2023 Feb.: One paper got accepted by Neurocomputing 2023!
2023 Feb.: Served as a reviewer for ICML 2023.
2023 Jan.: Served as a reviewer for CVPR 2023.
2022 Sept.: Our work ESOL got accepted by NeurIPS 2022!
2022 Feb.: Our work PPL got accepted by TMM 2022!

Publications

Weakly Supervised Semantic Segmentation via Self-Supervised Destruction Learning

Jinlong Li, Zequn Jie, Xu Wang, Yu Zhou, Lin Ma, Jianming Jiang
Neurocomputing (NEUCOM), 2023.

Abstract: In this paper, we propose a novel “destruction learning” method via self-supervised manner, producing the CAM attention maps better covering the whole object rather than only the most discriminative regions as previous approaches. Region destruction mechanism is proposed to deliberately “destruct” the global structure in both mid-level and low-level feature learning following jigsaw puzzle operation, for better local feature extraction of the classification network.

Expansion and Shrinkage of Localization for Weakly-Supervised Semantic Segmentation

Jinlong Li, Zequn Jie, Xu Wang, Xiaolin Wei, Lin Ma
Neural Information Processing Systems (NeurIPS), Spotlight (1.7%) 2022.
[ArXiv Link] [Code]

Abstract: We propose a new training pipeline to alleviate the partial localization issue of the CAM in Weakly-supervised image semantic segmentation, ESOL, in a Divide-and-Conquer manner.

Weakly Supervised Semantic Segmentation via Progressive Patch Learning

Jinlong Li, Zequn Jie, Xu Wang, Yu Zhou, Xiaolin Wei, Lin Ma
IEEE Transactions on Multimedia (TMM), 2022.
[ArXiv Link] [IEEE Trans Link] [Code]

Abstract: We propose a new training pipeline to alleviate the partial localization issue of the CAM in Weakly-supervised image semantic segmentation, PPL, in an iterative training manner.

Experience

Research Scientist, MeiTuan Inc.

2022.07 - 2023.10
Intern, MeiTuan Inc.

2020.12 - 2022.04
Image Algorithm Engineer, Reetoo Biotech.

2017.07 - 2019.04

Services

Journal Reviewer:

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

Conference Reviewer:

International Conference on Learning Representations (ICLR)
Neural Information Processing Systems (NeurIPS)
International Conference on Machine Learning (ICML)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
International Conference on Computer Vision (ICCV)
European Conference on Computer Vision (ECCV)

Awards

Outstanding student paper award of SZU, 2023