Training separate NIMA-style models on depth, saliency, and blur versions of AVA shows saliency carries the most signal among non-RGB modalities, while the standard 5.0 threshold inflates baselines to above 70 percent.
Datasets The dataset we chose to apply our methods to is the most widely used image aesthetics dataset; A V A or Aesthetic Vi- sual Analysis [6]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
REJECT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Explaining Automatic Image Assessment
Training separate NIMA-style models on depth, saliency, and blur versions of AVA shows saliency carries the most signal among non-RGB modalities, while the standard 5.0 threshold inflates baselines to above 70 percent.