What question did this study set out to answer?

This research aims to create a detailed dataset for analyzing the film techniques of Akira Kurosawa.

February 21, 2026Journal of Open Humanities Data0 citationsOpen Access

Kurosawa Shot-Level Dataset: Shot Boundaries, Color, and Face-Derived Shot-Scale Features for 30 Films

Key Points

This research aims to create a detailed dataset for analyzing the film techniques of Akira Kurosawa.
Developed dataset using automated shot segmentation techniques.
Extracted visual features including shot boundaries, face statistics, and color metrics.
Analyzed 30 films spanning from 1943 to 1993.
Established structured measurements for shot boundaries and durations.
Provided face-derived shot-scale proxies for in-depth analysis.
Documented color evolution metrics highlighting transitions from black-and-white to color.

Abstract

This paper presents a quality-flagged derived dataset for Akira Kurosawa’s films, containing structured measurements obtained from automated shot segmentation and visual feature extraction. The dataset covers 30 films (1943–1993). It includes shot boundaries and duration statistics (TransNet v2), intra-shot face statistics and face-derived shot-scale proxies (RetinaFace), and color metrics (brightness, warmth, saturation). The dataset supports cross-film comparisons of Kurosawa’s style, the analysis of color evolution from black-and-white to color cinema, and quantitative research in digital humanities and cinemetrics.

AIに質問

Bookmark

View Full Paper