1. IQA
  2. No-reference Image Quality Assessment
  3. Aesthetics

图像质量评估(Image Quality Assessment, IQA)即评估图像的质量。

以下是最近一些方法及其代码。

IQA

  • (ACM MM 2017, TMM 2019) No reference image quality assessment based Semantic Feature Aggregation

  • (CVPR 2018) Blind Predicting Similar Quality Map for Image Quality Assessment

  • (TIP 2018) NIMA: Neural Image Assessment



  • (TIP 2018) Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment

  • (ICCV 2017) RankIQA: Learning from Rankings for No-reference Image Quality Assessment

No-reference Image Quality Assessment

  • (CVPR 2018) Hallucinated-IQA: No-reference Image Quality Assessment via Adversarial Learning

  • (BMVC 2018) Self-supervised Deep Multiple Choice Learning Network for Blind Image Quality Assessment

  • (CVPR 2014) Convolutional Neural Networks for No-Reference Image Quality Assessment

  • (TIP 2012) No Reference Image Quality Assessment in the Spatial Domain

Aesthetics

  • (CVPR 2019) Effective Aesthetics Prediction with Multi-level Spatially Pooled Features

  • (PRIA 2019) Personalised aesthetics with residual adapters