4K & HD Vertical Live-Action Video Dataset for AI Training
Access 14,904 professionally filmed vertical live-action clips across 4K and Full HD resolution tiers. With approximately 75.7 hours of portrait-oriented footage, the collection is built for mixed-resolution mobile-video training, high-resolution vertical evaluation, and generative AI workflows.
Varied people, activities, environments, and camera conditions provide realistic portrait-oriented coverage, while structured metadata supports resolution-aware sampling, portrait-video retrieval, multimodal learning, and controlled evaluation.
Dataset Preview
Representative previews showing people, activities, environments, and camera conditions across the vertical 4K and Full HD corpus.
The examples illustrate portrait framing and scene variation across both resolution tiers, including portrait composition, movement, and environmental context.
Key Highlights
- • Mixed 4K and Full HD tiers for resolution-aware portrait workflows
- • Portrait-oriented framing for mobile and short-form video systems
- • 4K detail for high-resolution vertical model and output assessment
- • Structured resolution metadata for controlled sampling and benchmark splits
- • Broad coverage of portrait subjects, actions, environments, and camera conditions
Metadata Fields
Technical Specifications
Dataset Type
Mixed-Resolution Live-Action Corpus: 4K + HD Vertical
Primary Technical Intent
High-resolution portrait video, mobile generation, and resolution-aware vertical training
Content Type
Professionally filmed vertical live-action video
Clip Count
14,904 clips
Total Duration
75.7 hours
Average Clip Length
Approximately 18.3 seconds
Resolution
4K: 2160 × 4096
Full HD: 1080 × 1920
Resolution Distribution
Counts by 4K and Full HD tier can be reported for the selected delivery
Formats
4K + HD
Codec
H.264
4K Bitrate
Approximately 87–93 Mbps
HD Bitrate
Approximately 22 Mbps
Container
MP4
Orientation
Vertical
Aspect Ratio Mixed portrait formats
9:16 Full HD; 135:256 4K where available
Dialogue / Spoken Language
N/A, no dialogue or voiceover
Subtitles / Scripts
N/A, no subtitles or scripts
Licensing, Documentation, and Delivery
Wavebreak Media can provide applicable licensing, provenance, and model or property release information for the video, with dataset delivery organized around agreed file, metadata, and technical requirements. See our Dataset Licensing & Compliance and Dataset Delivery & Security pages for details on rights documentation, packaging, and secure transfer.
Need the Full Dataset?
Request a portrait-video sample across the 4K and Full HD tiers to evaluate framing, metadata, resolution coverage, and fit for mobile generation or high-resolution vertical models.
Related Datasets
HD Vertical Live-Action Video Dataset for AI Training
A standardized 1080 × 1920 corpus of 78,205 vertical clips for short-form pipelines, mobile retrieval, classification, and repeatable AI evaluation.
View Dataset Video Datasets for AI Training4K & HD Horizontal Live-Action Video Dataset for AI Training
A licensed mixed-resolution corpus of 287,265 horizontal 4K/UHD and Full HD clips for resolution-robust AI training, high-resolution evaluation, and retrieval.
View Dataset Video Datasets for AI TrainingHD Horizontal Live-Action Video Dataset for AI Training
A standardized 16:9 corpus of 73,328 Full HD 1920 × 1080 live-action clips for scalable decoding, retrieval, classification, and repeatable AI evaluation.
View Dataset