When was that made?

TL;DR

This paper develops deep learning techniques to estimate the creation date of objects, aiding historical research and data organization, by leveraging existing and new neural network models trained on large datasets.

Contribution

It introduces a novel approach combining existing features and fine-tuned networks for temporal estimation, along with new datasets and comprehensive analysis.

Findings

Outperforms previous state-of-the-art methods

Creates two large datasets of dated clothing items

Provides insights into what neural networks learn for temporal estimation

Abstract

In this paper, we explore deep learning methods for estimating when objects were made. Automatic methods for this task could potentially be useful for historians, collectors, or any individual interested in estimating when their artifact was created. Direct applications include large-scale data organization or retrieval. Toward this goal, we utilize features from existing deep networks and also fine-tune new networks for temporal estimation. In addition, we create two new datasets of 67,771 dated clothing items from Flickr and museum collections. Our method outperforms both a color-based baseline and previous state of the art methods for temporal estimation. We also provide several analyses of what our networks have learned, and demonstrate applications to identifying temporal inspiration in fashion collections.