C57_S01: AI-Generated and Original Video Scenes for Style Analysis

A doctoral dissertation project provides scene-level video materials for the film 'Cléo from 5 to 7'. The dataset includes RAW and PRE-processed videos, prompt files, generation settings, and metadata for comparative analysis of original and AI-generated content. It was authored by Jinyu Kim and last updated on June 4, 2026.

Use Cases

• Comparative visual style analysis based on original and AI-generated video pairs
• Benchmarking generative AI video models against a canonical film scene
• Studying the reconfiguration of cinematic techniques like those in the French New Wave
• Analyzing the effect of standardized preprocessing (720p, 24 fps, 8 seconds) on AI video output

Strengths

• Includes both RAW generated videos and standardized PRE-processed versions for controlled analysis
• Provides specific generation settings and prompt files, enabling reproducibility
• Focuses on a single, well-documented scene (C57_S01) from a canonical 1962 film
• Dataset is 44.6 MB in size, facilitating relatively quick download and inspection

Limitations

• Row count is unknown, which may limit suitability assessment
• Column-level documentation is absent; field semantics must be inferred after download
• Audio is excluded from the analysis, limiting the scope of multimodal study

Provenance

Source

figshare

Collection Method

Created as part of a doctoral dissertation project on film style reconfiguration through generative AI.

Time Range

Scene is from the 1962 film 'Cléo de 5 à 7'.

Freshness

Last updated 2026-06-04 20:58:29; freshness should be verified