CAL_S01: AI-Generated Video Style Analysis for The Cabinet of Dr. Caligari

Jinyu Kim's doctoral dissertation project provides scene-level video materials for a comparative analysis of original and AI-generated content. The dataset includes RAW and PRE videos, prompt files, generation settings, and metadata for the 1920 German Expressionist film 'The Cabinet of Dr. Caligari'. Materials were last updated on 2026-06-04 and are shared under a CC-BY-4.0 license.

Use Cases

• Comparative visual analysis of original and AI-generated film scenes based on the provided RAW and PRE video files.
• Studying the reconfiguration of film styles through generative AI models based on the project's stated research focus.
• Analyzing the relationship between text prompts and generated video output based on the included prompt files and generation settings.
• Benchmarking AI video generation models against a classic film source based on the standardized 720p, 24 fps, 8-second PRE video clips.

Strengths

• Includes both RAW generated videos and standardized PRE videos (720p, 24 fps, 8 seconds) for consistent analysis.
• Contains auxiliary data like prompt files and generation settings, which likely provide context for the AI video creation process.
• Focuses on a well-defined, historically significant film ('The Cabinet of Dr. Caligari', 1920) with a clear artistic style (German Expressionism).

Limitations

• Row count and specific column-level documentation are unknown, which may limit suitability assessment.
• Audio is excluded from the analysis, limiting the dataset's utility for multimodal audio-visual research.
• The dataset is small (43.8 MB), which may constrain the scale of analysis possible.

Provenance

Source

Jinyu Kim, via figshare.

Collection Method

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

Time Range

Source film is from 1920; dataset creation date is unspecified.

Freshness

Last updated 2026-06-04 21:00:25; freshness should be verified.