Dingquan Li

I am a Postdoc at Peng Cheng Laboratory (PCL), where I am advised by Prof. Ge Li. I did my PhD at School of Mathematical Sciences, BICMR at Peking University, where I was advised by Prof. Ming Jiang and Prof. Tingting Jiang. During this period, I was also a research assistant in the VIE group at NELVT of Peking University. Before coming to Peking University in 2015, I did my bachelors at Nankai University.

Email  /  CV;CV_cn  /  Google Scholar  /  Github  /  ResearchGate  /  Linkedin

profile photo
Research

My main interest lies in image, video, and point cloud processing, including:

  • Point Cloud Quality Assessment
  • Image/Video Quality Assessment
  • Perceptual Optimization
Unified Quality Assessment of In-the-Wild Videos with Mixed Datasets Training
Dingquan Li, Tingting Jiang, Ming Jiang
International Journal of Computer Vision (IJCV) Special Issue on Computer Vision in the Wild, 2021.   (SCI JCR Q1, IF=5.698; CCF A)
code / bibtex

A mixed datasets training strategy for training a single unified VQA model with multiple datasets

Which Has Better Visual Quality: The Clear Blue Sky or a Blurry Animal?
Dingquan Li, Tingting Jiang, Weisi Lin, Ming Jiang
IEEE Transactions on Multimedia (TMM), 2019   (SCI JCR Q1, IF=5.452; CCF B)
slides / code / bibtex

Most conventional objective metrics prefer blurry animals (relatively complex visual content) over clear the blue sky (simple content), contradicting with human perception: content-aware features help ...

Norm-in-Norm Loss with Faster Convergence and Better Performance for Image Quality Assessment
Dingquan Li, Tingting Jiang, Ming Jiang
ACM International Conference on Multimedia (MM), 2020   (Oral, CCF A)
video / poster / code / bibtex

Normalization-embedded loss is conducive to the faster convergence and better performance of the IQA model.

Quality Assessment of In-the-Wild Videos
Dingquan Li, Tingting Jiang, Ming Jiang
ACM International Conference on Multimedia (MM), 2019   (Oral, CCF A)
slides / poster / code / bibtex

Content dependency and temporal-memory effects of HVS are considered in the design of NR-VQA models.

Exploiting High-Level Semantics for No-Reference Image Quality Assessment of Realistic Blur Images
Dingquan Li, Tingting Jiang, Ming Jiang
ACM International Conference on Multimedia (MM), 2017   (CCF A)
slides / poster / code / bibtex

High-level semantic features extracted from pre-trained image classification models help NR-IQA.

Blur-Specific No-Reference Image Quality Assessment: A Classification and Review of Representative Methods
Dingquan Li, Tingting Jiang
Proceedings of the International Conference on Sensing and Imaging, 2019   (Invited Chapter)
bibtex

Recent Advances and Challenges in Video Quality Assessment
Dingquan Li, Tingting Jiang, Ming Jiang
ZTE Communications, 2019   (Invited Paper)
bibtex

Quality Assessment for Tone-Mapped HDR Images Using Multi-Scale and Multi-Layer Information
Qin He, Dingquan Li, Tingting Jiang, Ming Jiang
IEEE International Conference on Multimedia & Expo Workshop (ICMEw), 2018  
code / bibtex

Academic Services
Reviewer for TIP, TMM, TCSVT, CVPR, ICCV, ACM MM, IJCAI, etc.

Secondary Reviewer for AAAI, etc.


Template credit to Jon Barron