InnerEye: A Tale on Images Filtered Using Instagram Filters - How Do We Interact with them and How Can We Automatically Identify the Extent of Filtering?

Gazi Abdur Rakib, Rudaiba Adnin, Shekh Ahammed Adnan Bashir, Chashi Mahiul Islam, Abir Mohammad Turza, Saad Manzur, Monowar Anjum Rashik, Abdus Salam Azad, Tusher Chakraborty, Sydur Rahaman, Muhammad Rayhan Shikder, Syed Ishtiaque Ahmed, A. B. M. Alim Al Islam

Abstract: Even though digitally filtered images are taking over the Internet for their aesthetic appeal, general people often feel betrayed if they are dealt with filtered images. Our study, comprising a series of structured surveys over images filtered using Instagram filters, reveals that different people perceive the filtered images differently. However, people have a common need for an automated tool to help them distinguish between original and filtered images. Accordingly, we develop an automated tool named ‘InnerEye’, which is capable of identifying how far an image is filtered or not. InnerEye utilizes a novel analytical design of a Neural Network Model that learns from a diverse set of images filtered using Instagram filters. Rigorous objective and subjective evaluations confirm the efficacy of InnerEye in identifying the extent of filtering in the images.