NNCD-IQA: A new neural networks based compressed database for image quality assessment

Objective and subjective quality assessment is still a challenging problem in various image processing tasks. For instance, the context of compression, most conducted studies have focused on datasets encoded using standard algorithms such as JPEG JPEG2000. In this paper, we propose to further investigate issue presence neural networks-based compressed images. More precisely, new database images has been firstly built JPEG2000 well four recent networks based coding schemes. Then, experiments are performed obtain mean opinion scores generated distorted Finally, an extensive evaluation analysis objective metrics achieved. addition conventional machine learning metrics, considered different deep models, which trained our database. The with its associated learned models publicly available at https://github.com/zakopz/NNCD-IQA-Database . obtained results show interest