← Back to Publications

The Focus–Aspect–Value Model for Predicting Subjective Visual Attributes

P. Blandfort, T. Karayil, J. Hees, A. Dengel

International Journal of Multimedia Information Retrieval (IJMIR)||DOI: 10.1007/s13735-019-00188-5
Visual AttributesSubjectivityDeep Learning

Abstract

We propose the Focus–Aspect–Value model for structuring the process of capturing subjectivity in image processing, addressing the challenge of predicting subjective visual attributes from images.

Overview

Subjective visual interpretation is a challenging yet important topic in computer vision. Many approaches reduce this problem to the prediction of adjective- or attribute-labels from images.

The FAV Model

We propose the Focus–Aspect–Value (FAV) model to structure the process of capturing subjectivity in image processing. This model provides a principled framework for understanding how subjective attributes relate to visual content.