geekyutao / GraphIQA

GraphIQA: Learning Distortion Graph Representations for Blind Image Quality Assessment
Apache License 2.0
42 stars 8 forks source link

Enquire about the database splitting - 60% training, 20% validation, and 20% testing #3

Closed SuperBruceJia closed 2 years ago

SuperBruceJia commented 2 years ago

Dear,

Thanks a lot for your great work!

Could you share your experimental results (SRCC and PLCC) (Database: LIVE, CSIQ, TID2013, and KADID-10k) by splitting the dataset into 60% training, 20% validation, and 20% testing?

Thanks in advance and have a nice day!

Best,

Shuyue Feb 8, 2022

sunsm-2020 commented 2 years ago

The experimental results are as follows: LIVE: SRCC 0.9684 PLCC 0.9704 CSIQ: SRCC 0.9203 PLCC 0.9376 KADID-10k: SRCC 0.8846 PLCC 0.8840