AI & Machine Learning
Vintage Film AI Training Dataset
License a large-scale dataset of vintage 8mm and Super 8 archival footage for computer vision, generative AI, and temporal analysis research. Over 162,603 digitized clips with structured metadata spanning the 1930s–1980s.
Key Stats
Dataset Contents
Each clip record includes the following structured metadata fields:
- Title — descriptive clip title
- Description — detailed narrative description
- Keywords — searchable keyword tags
- Shot Year — estimated year the footage was captured
- Shot Decade — decade classification (e.g. 1950s, 1960s)
- Duration — clip length in milliseconds
- Format Tags — film format (8mm, Super 8)
- Location — city, region, and country
- Geo Tokens — tokenized geographic identifiers
- Copyright — rights and ownership information
Metadata coverage across the full dataset:
Content Flags
Across the 118,435 clips with a completed rights review, each carries flags stored in the licensing catalog, so you can scope a subset to your compliance requirements before anything is signed:
- People visible — 82,319 clips
- Minors visible — 40,040 clips; subsets that exclude them are available on request
- Logos or trademarks — 16,002 clips
- Audible music — 1,955 clips
All footage was shot decades before generative models existed, so the corpus carries no synthetic contamination — and the reel-level chain of title lets you verify that clip by clip.
Film Format Breakdown
All footage in the collection was originally shot on analog film stock:
Decade Coverage
Distribution of clips by the decade in which they were shot:
Geographic Coverage
Footage spans 121 countries and 935+ cities. Top 15 countries by clip count:
Still Images
In addition to video clips, the dataset includes approximately 800,000 still frame extractions. These high-resolution images are derived from key frames across the archive and can be used independently for image classification, object detection, and visual similarity research.
Use Cases
Computer Vision
Train object detection, scene classification, and activity recognition models on authentic mid-century imagery.
Generative AI
Fine-tune video and image generation models to produce realistic vintage aesthetics including film grain, color shifts, and period artifacts.
Temporal Analysis
Study visual changes over decades across consistent geographic locations, fashion, architecture, and urban landscapes.
Multimodal Research
Pair rich textual metadata with visual content for vision-language model training and cross-modal retrieval.
Cultural Preservation
Develop AI tools for automated restoration, colorization, and cataloging of historical film archives.
Case Study: Paris, France 1947
To test the dataset on generative AI, we trained a model on roughly 2,500 still images pulled from a single location and era: Paris circa 1947. The output was hard to tell apart from real archival frames. It picked up the film grain, the color palette, and the period look.
Browse the source collection at 1947 Paris France to see the original training material. The trained model is open-sourced and available for download — try it yourself and see the results firsthand at CivitAI.
Access & Pricing
Standard Stockfilm clip licenses do not include AI-training rights. Dataset licenses are separate agreements that name the permitted uses, which keeps the chain of title clean for your legal review.
Evaluation sample
Free. 100 clips with complete metadata records, matched to your use case. Requests are reviewed before delivery.
Research license
From $4,900. Subsets up to 25,000 clips for non-commercial training and publication. Metadata delivered as CSV or JSON.
Commercial training license
From $0.95 per clip. Volume-tiered. Filter by decade, geography, subject, or content flags. Full-corpus and refresh terms are negotiated directly.
Live counts behind every claim on this page are published in the machine-readable datasheet, generated from the same catalog database that fulfills orders.
Frequently Asked Questions
Video clips are available as MP4 files. Metadata can be delivered as CSV, JSON, or via API access. Still frame extractions are available as high-resolution JPEG or PNG files.
Yes. We offer custom subsets filtered by decade, geography, format, or keyword. Contact us to discuss your specific requirements.
Research licenses cover non-commercial training and academic publication. Commercial licenses cover production use in AI products and services. Either way, training rights are granted explicitly in the dataset agreement — the standard per-clip license sold on this site does not include them.
All metadata fields are structured and machine-readable. Keywords are tokenized, locations are normalized, and temporal data follows consistent formatting for easy integration with ML pipelines.
Source film is scanned at 4K, which is what the per-clip licenses on this site deliver. Dataset deliveries can be 4K masters, HD mezzanines, or lower-resolution analysis proxies — most teams take HD or proxies to keep storage manageable. The grain, color shifts, and period artifacts survive at every resolution.
License This Dataset
Start with the free 100-clip evaluation sample, or go straight to scoping a subset. Tell us the use case and we respond with a concrete proposal: clip counts, delivery format, and price.
Request the Evaluation Sample Download the Datasheet (JSON)