CUSTOMVIDEOFIRST-PERSON
First-Person Activity Clips
Wearable-style first-person activity clips with task labels, step metadata, camera perspective, hands-visible review, environment fields, and optional narration for embodied AI evaluation.
CUSTOMVIDEOINSTRUCTIONAL
Procedural Demonstration Video Sets
Step-by-step demonstration video collection format for imitation learning, procedural AI, video-LLM grounding, and instruction-following evaluation.
Procedural Demonstration Video Sets are built for teams evaluating models that need to understand ordered actions, task progress, step completion, and instruction-to-action alignment. The format captures practical demonstrations from start to finish with step structure and metadata.
Structured collection
Procedural Demonstration Video Sets use defined practical tasks with clear start, middle, and completion conditions.
Metadata depth
The format supports alignment between video timestamps, step labels, narration transcript, object mentions, action labels, and…
Delivery-ready package
Includes video quality, step clarity, task completion reviews, plus dataset manifest, CSV catalog, consent status metadata, QA…
Request sample access
Share your target procedure types, step schema, camera setup, annotation needs, and evaluation goals. We’ll reply with a technical scope for the collection.
Key highlights
- Step-by-step task demonstrations with timestamped step labels.
- Structured task metadata for category, step count, completion status, duration, and camera setup.
- Optional narration transcript aligned to steps and video timestamps.
Dataset contents
Video capture
Procedural video files · Optional narration audio
Steps, transcripts, and labels
Timestamped step markers · Optional narration transcript
Capture and task metadata
Task category metadata · Step labels · Completion status · Camera setup metadata · Environment metadata · Duration metadata
Optional annotation layers
Optional instruction alignment labels
Technical specifications
Use cases
- Imitation learning evaluation
- Procedural AI testing
- Video-LLM step grounding
- Instruction following
- Task planning model evaluation
- Step segmentation
- Action sequence modeling
- Multimodal assistant testing
Dataset workflow
1Collection
Procedural Demonstration Video Sets use defined practical tasks with clear start, middle, and completion conditions. Contributors record task execution under capture instructions that preserve step order, visual clarity, and outcome evidence.
2Review
Video quality review · Step clarity review · Task completion review · Narration alignment review where scoped · Metadata review · Consent status review · Delivery approval
3Delivery
MP4 video files · Step metadata JSON · Task metadata CSV · Narration transcript where scoped · Annotation files where scoped · Dataset manifest JSON · CSV file catalog · QA summary · Consent status metadata · Delivery notes
Quality assurance
Related dataset formats
CUSTOMVIDEODEXTEROUS TASKS
Hand and Object Interaction Clips
Hand-object interaction clips with action verbs, object categories, camera angle metadata, contact-event markers, object-state labels, and QA fields for robotic manipulation evaluation.
CUSTOMVIDEOENVIRONMENT DIVERSITY
Multi-Environment Activity Clips
Activity video sets captured across varied real-world environments, with task labels, environment metadata, camera position fields, lighting conditions, duration, and QA status for activity recognition and embodied AI evaluation.
CUSTOMSCREENNARRATED WORKFLOWS
Guided Software Walkthrough Sessions
Narrated software walkthroughs with screen recordings, aligned narration transcripts, step markers, app context, cursor flow, and task metadata for procedural software understanding.
CUSTOMMULTIMODALPAIRED DATA
Multimodal Paired Capture Sets
Paired multimodal datasets linking video, audio, screen, text, or image-derived records through shared file IDs, cross-modal manifests, metadata schemas, and QA fields.
CUSTOMANY MODALITYBUYER-SCOPED
Buyer-Scoped Custom Collection
Fully scoped custom data collection with buyer-defined modality, schema, metadata fields, annotation layers, QA criteria, manifest structure, delivery package, and collection rules.