A picture is worth a thousand words: Image analysis for the Digital Humanities

Workshop ID:

WT-04

Workshop Title:

A picture is worth a thousand words: Image analysis for the Digital Humanities

Organizers:

Stuart James, Mathieu Aubry, Nanne van Noord, Noa Garcia, Leonardo Impett

Time (in JST and UTC):

July 25 18:00-21:30 (JST)

July 25 9:00-12:30 (UTC)

Maximum number of participants:

100

Description:

In this tutorial, we aim to provide insights into how Computer Vision (CV) methods can be used to enhance study within the Digital Humanities (DH). In contrast to the digitization use of CV, we focus on the interpretation of the content that can be used for distant reading. We cover four State-of-the-Art research problems within CV that are being applied to both art and cultural heritage visual data, then considering how CV methods are being applied to answer DH questions. The tutorial brings together four CV and one DH academics with extensive experience in applying CV to DH.

Aim of the workshop/tutorial:

In this tutorial, we aim to provide insights into how Computer Vision (CV) methods can be used to enhance study within the Digital Humanities (DH). In contrast to the digitization use of CV, we focus on the interpretation of the content that can be used for distant reading. We cover four State-of-the-Art research problems within CV that are being applied to both art and cultural heritage visual data, then considering how CV methods are being applied to answer DH questions. The tutorial brings together four CV and one DH academics with extensive experience in applying CV to DH.

Outline:

The workshop is broken into five sections:

Part 1: Retrieval and Knowledge Graphs (Stuart James)

Part 2: Content-based analysis (Mathieu Aubry)

Part 3: Multi-Task Learning (Nanne van Noord)

Part 4: Automatic interpretation (Noa Garcia)

Part 5: Using Computer Vision within humanities research (Leonardo Impett)