A visual complexity dataset that includes both single-score data as well as two-dimensional annotations of simple and complex regions. Two datasets are included; a collection upright scenes with the aforementioned human data, and a collection of inverted scene images with corresponding complexity ratings.
Upright scene images, complexity scores, and complexity maps. For visualisation purposes, complexity maps normalised by annotation count are available. For data processing purposes, the base maps should be used.
Inverted scene images, complexity scores and complexity maps for these inverted scene images.