Automatic Image Transformation for Inducing Affect

TL;DR

This paper introduces an automatic image transformation technique that uses deep learning and emotion transfer to modify images for inducing specific emotional responses, without needing a target image.

Contribution

It presents a novel emotion transfer algorithm that leverages deep features and emotion distributions to enhance images' emotional impact, handling diverse photographs.

Findings

User study confirms effectiveness of the method.

Method outperforms previous approaches in diversity of photographs.

Identifies limitations of color-transfer in emotion assignment.

Abstract

Current image transformation and recoloring algorithms try to introduce artistic effects in the photographed images, based on user input of target image(s) or selection of pre-designed filters. These manipulations, although intended to enhance the impact of an image on the viewer, do not include the option of image transformation by specifying the affect information. In this paper we present an automatic image-transformation method that transforms the source image such that it can induce an emotional affect on the viewer, as desired by the user. Our proposed novel image emotion transfer algorithm does not require a user-specified target image. The proposed algorithm uses features extracted from top layers of deep convolutional neural network and the user-specified emotion distribution to select multiple target images from an image database for color transformation, such that the resultant image has desired emotional impact. Our method can handle more diverse set of photographs than the previous methods. We conducted a detailed user study showing the effectiveness of our proposed method. A discussion and reasoning of failure cases has also been provided, indicating inherent limitation of color-transfer based methods in the use of emotion assignment. Project Page: http://im.itu.edu.pk/affective-image-transfer/