“Distant viewing” in digital humanities is an analytical approach adapted from Franco Moretti’s concept of “distant reading.” While distant reading involves analyzing large collections of texts computationally to identify patterns and trends, distant viewing applies similar principles to visual media like films, television shows, photographs, and artwork.
The term was notably developed by Taylor Arnold and Lauren Tilton, who define distant viewing as a computational approach to analyzing large collections of visual media that combines:
- Computer vision techniques (for detecting and classifying objects, faces, scenes, text, etc.)
- Statistical analysis of visual patterns
- Cultural and historical contextualization
Key aspects of distant viewing include:
- Automated analysis of visual features like color schemes, composition, camera angles, and scene transitions
- Detection and tracking of recurring visual elements across large collections
- Quantitative analysis of visual patterns and their changes over time
- Integration of visual analysis with other forms of data like metadata, scripts, or historical records
For example, scholars might use distant viewing to:
- Analyze how gender representation in television shows has evolved over decades
- Study changing cinematographic techniques across film history
- Identify patterns in historical photograph collections
- Track the evolution of visual styles in artwork or advertising
The approach allows researchers to analyze visual media at scales that would be impossible through traditional close analysis, while still maintaining connection to cultural and historical context.