Discovering beautiful attributes for aesthetic image analysis
This work addresses the need for interpretable aesthetic analysis in photography and image processing, though it is incremental as it builds on existing datasets and methods.
The paper tackles the problem of achieving both accuracy and interpretability in aesthetic image analysis by proposing to automatically discover and learn nameable visual attributes from the AVA dataset, which contains over 250,000 images with aesthetic scores and comments. The result is that these learned attributes are successfully applied to aesthetic quality prediction, image tagging, and retrieval.
Aesthetic image analysis is the study and assessment of the aesthetic properties of images. Current computational approaches to aesthetic image analysis either provide accurate or interpretable results. To obtain both accuracy and interpretability by humans, we advocate the use of learned and nameable visual attributes as mid-level features. For this purpose, we propose to discover and learn the visual appearance of attributes automatically, using a recently introduced database, called AVA, which contains more than 250,000 images together with their aesthetic scores and textual comments given by photography enthusiasts. We provide a detailed analysis of these annotations as well as the context in which they were given. We then describe how these three key components of AVA - images, scores, and comments - can be effectively leveraged to learn visual attributes. Lastly, we show that these learned attributes can be successfully used in three applications: aesthetic quality prediction, image tagging and retrieval.