Seeing Behind the Camera: Identifying the Authorship of a Photograph

TL;DR

This paper presents a new approach to identify the photographer behind a photograph using a large dataset and deep learning, revealing that high-level features significantly improve identification accuracy.

Contribution

The study introduces the novel problem of photographer identification, creates a large dataset, and develops a deep CNN that outperforms traditional features in this task.

Findings

High-level features outperform low-level features in photographer identification

Deep CNN achieves high accuracy in distinguishing photographers

Qualitative analysis reveals distinctive photographic styles

Abstract

We introduce the novel problem of identifying the photographer behind a photograph. To explore the feasibility of current computer vision techniques to address this problem, we created a new dataset of over 180,000 images taken by 41 well-known photographers. Using this dataset, we examined the effectiveness of a variety of features (low and high-level, including CNN features) at identifying the photographer. We also trained a new deep convolutional neural network for this task. Our results show that high-level features greatly outperform low-level features. We provide qualitative results using these learned models that give insight into our method's ability to distinguish between photographers, and allow us to draw interesting conclusions about what specific photographers shoot. We also demonstrate two applications of our method.