Brain-Inspired Deep Networks for Image Aesthetics Assessment
This work addresses image aesthetics assessment, a domain-specific problem for applications like photography and art, with incremental contributions through a novel network design and data augmentation.
The paper tackles the subjective challenge of image aesthetics assessment by proposing Brain-Inspired Deep Networks (BDN), which learn attributes through parallel supervised pathways and synthesize them into ratings, including predicting rating distributions; experimental results on the AVA dataset show significant performance improvements compared to state-of-the-art models.
Image aesthetics assessment has been challenging due to its subjective nature. Inspired by the scientific advances in the human visual perception and neuroaesthetics, we design Brain-Inspired Deep Networks (BDN) for this task. BDN first learns attributes through the parallel supervised pathways, on a variety of selected feature dimensions. A high-level synthesis network is trained to associate and transform those attributes into the overall aesthetics rating. We then extend BDN to predicting the distribution of human ratings, since aesthetics ratings are often subjective. Another highlight is our first-of-its-kind study of label-preserving transformations in the context of aesthetics assessment, which leads to an effective data augmentation approach. Experimental results on the AVA dataset show that our biological inspired and task-specific BDN model gains significantly performance improvement, compared to other state-of-the-art models with the same or higher parameter capacity.