DeepRare: Generic Unsupervised Visual Attention Models

Abstract

Human visual system is modeled in engineering field providing feature-engineered methods which detect contrasted/surprising/unusual data into images. This data is "interesting" for humans and leads to numerous applications. Deep learning (DNNs) drastically improved the algorithms efficiency on the main benchmark datasets. However, DNN-based models are counter-intuitive: surprising or unusual data is by definition difficult to learn because of its low occurrence probability. In reality, DNN-based models mainly learn top-down features such as faces, text, people, or animals which usually attract human attention, but they have low efficiency in extracting surprising or unusual data in the images.

Using artificial intelligence for predictive eye-tracking analysis to evaluate photographs

“…In recent years, artificial intelligence (AI) has proven to be useful in many fields, and photography is no exception (Kong et al, 2022). There are well-known and commonly used cases of computational photography and image recognition and we can even see some algorithms capable of generating a photographic-looking images based on just a few keywords (OpenAI, 2022;Dechterenko & Lukavsky, 2016).…”